Ornithologists at the Cornell Lab of Ornithology have unveiled the most comprehensive evolutionary tree of birds to date. This groundbreaking research reveals unexpected relationships and serves as a fascinating illustration for bird enthusiasts. Explore the World Bird Lineage Explorer, where you can track lineage and evolutionary milestones.
European bee-eater (Merops apiaster). Image credit: Rashuli / CC BY 2.0.
Understanding the phylogeny of birds is crucial for advancing bird research.
With over 11,000 bird species worldwide, consolidating phylogenetic trees into a singular, updated resource has posed significant challenges for ornithologists.
The Birds of the World Phylogeny Explorer directly addresses these challenges, remaining current with the latest scientific discoveries.
“This tool combines centuries of avian research with advanced computational tools, creating a captivating interactive resource that narrates the story of bird evolution,” stated Dr. Elliott Miller, a researcher with the American Bird Conservancy.
“New evolutionary relationships are constantly being discovered. We release annual updates to our phylogenetic tools, ensuring our datasets align with the latest taxonomy,” he added.
“This tool holds immense value for the scientific community,” remarked Dr. Pam Rasmussen from the Cornell Lab of Ornithology.
“The complete tree of bird life, built on cutting-edge phylogenetic research, is now a downloadable, interactive dataset from Birds of the World, encouraging further inquiry and exploration.”
“This evolutionary tree provides crucial insights into how evolutionary history has shaped traits such as beak shape, wing length, foraging behaviors, and habitat preferences in birds.”
“Bird lovers will appreciate the personalized features of the Birds of the World Phylogeny Explorer,” Dr. Marshall Iliff noted, also from the Cornell Lab of Ornithology.
“By logging into the platform, birders can visualize the diversity of their eBird species list, diving deep into bird history across orders, families, and genera, thus revealing noteworthy evolutionary patterns.”
“For birdwatchers, their lifetime list transforms into a personal journey through evolutionary history, highlighting how each species fits into the broader narrative of avian evolution.”
“Users are sure to encounter surprising revelations. For instance, why does the North American woodpecker closely resemble other woodpeckers yet belong to a different lineage?”
“Or why are peregrine falcons fierce hunters like hawks and eagles, even though they originate from a separate branch of the family tree?”
“Solving these taxonomic enigmas can become a lifelong pursuit for anyone deeply passionate about birds.”
Archaeologists have made a groundbreaking discovery, unearthing the “oldest known hand-held wooden tool” at a Middle Pleistocene site in Marathusa 1, Greece.
Impression of a Marathusa 1 female artist crafting a digging stick using small stone tools from an alder trunk. Image credit: G. Prieto / K. Harvati.
According to Professor Katerina Herberty from the University of Tübingen, “The Middle Pleistocene was crucial for human evolution, marking a period when complex behaviors emerged.”
“This era also showcases the earliest reliable evidence of the targeted use of plants for technological purposes.”
The 430,000-year-old wooden tools discovered at the Marathusa 1 site, led by Professor Harbati and his team, consist of worked alder trunks and small willow/poplar artifacts.
The primary tool is made from alder wood (Alnus sp.) and features engraving marks along with associated stop and chop marks, indicating intentional shaping.
This approximately 81 cm long artifact displays signs of usage consistent with a multifunctional rod likely employed for paleolakeshore excavation.
The second tool, a small piece of willow/poplar (Salix sp./Populus sp.), measures 5.7 cm and exhibits signs of rounding.
This object shows two signs of potential processing, suggesting that growth rings have been removed from one end.
Researchers hypothesize that this small wooden tool’s function remains uncertain but may have been utilized for modifying stone tools.
Alongside these wooden tools, scientists uncovered butchered remains of an elephant with straight tusks (Paleoloxodon Antique), as well as stone artifacts and processed bones.
Dr. Annemieke Milks, a researcher at the University of Reading, states, “Unlike stone artifacts, wooden objects need special conditions to survive over long durations.”
“We meticulously examined all tree remains, analyzing the surfaces under a microscope.”
“Our findings revealed clear evidence of cutting and carving on these two objects, strongly indicating that early humans intentionally shaped them.”
A multifunctional digging stick (top) and small wooden tools (bottom) from the Marathusa 1 site in Greece. Image credit: D. Michailidis / N. Thompson / K. Harvati.
Additionally, researchers found a large fragment of an alder trunk exhibiting deep carved stripes, interpreted as fossilized claw marks from a large carnivore. This suggests potential competition between early humans and carnivores at this site.
Evidence of cuts and damage on the elephant remains indicate that early hominins had access to the carcass, while gnawing marks reveal subsequent carnivorous activity.
Dr. Milks added, “Previous discoveries of ancient wooden tools have occurred in countries such as Britain, Zambia, Germany, and China, comprising weapons, digging sticks, and tool handles.” However, she noted that these finds date newer than the Marathusa 1 artifacts.
“The only evidence of ancient wood used by humans, dating to around 476,000 years ago, comes from the Kalambo Falls site in Zambia, where the wood served as structural material rather than tools.”
“We have now identified the oldest known wooden tools and the first of their kind from southeastern Europe,” emphasized Professor Herberty.
“This discovery highlights the exceptional conservation conditions at the Marathusa 1 site.”
“The concurrent evidence of human activity and large carnivores in the vicinity of the butchered elephant indicates a competitive dynamic between them.”
Details of these findings are published in Proceedings of the National Academy of Sciences.
_____
A. Chemilux et al. 2026. The earliest evidence of human use of wooden hand tools, discovered at Marathusa 1 (Greece). PNAS 123 (6): e2515479123; doi: 10.1073/pnas.25154791
Technological advancements in Africa and Western Europe during the late Middle Pleistocene highlight the intricate behaviors of hominin groups. Contrarily, East Asian human technology has long been perceived as lacking innovation. Recent archaeological findings at the Xigou site in Henan province, China, reveal remarkable evidence of technological innovations dating back between 160,000 and 72,000 years, illustrating over 90,000 years of sophisticated technological behavior through detailed technological, typological, and functional analyses.
Artist’s restoration of the Nishimizo utensil holder. Image credit: Hulk Yuan, IVPP.
“For decades, researchers have posited that, while Africa and Western Europe exhibited significant technological growth, East Asians relied on simpler and more traditional stone tool techniques,” noted Dr. Shisia Yang from the Institute of Vertebrate Paleontology and Paleoanthropology.
In recent findings, Dr. Yang and colleagues reveal that, during a time when numerous large-brained hominins coexisted in China, the hominins in this region displayed far greater inventiveness and adaptability than previously assumed, including species such as Homolonghi, Homo juruensis, and potentially Homo sapiens.
“The discovery at Xigou challenges the notion that early human populations in China were inherently conservative over time,” emphasized Professor Michael Petraglia from Griffith University.
“In-depth analyses indicate that the early inhabitants utilized advanced stone tool-making techniques to create small flakes and multifunctional tools,” he added.
Notably, the site revealed handled stone tools, marking the earliest known evidence of composite tools in East Asia.
These tools, which integrated stone components with handles and shafts, demonstrate exceptional planning, skilled craftsmanship, and knowledge of how to enhance tool functionality.
“Their existence underscores the behavioral flexibility and ingenuity of the Nishigou hominids,” Dr. Jiang Ping Yue, also affiliated with the Institute of Vertebrate Paleontology and Paleoanthropology, remarked.
The geological formations at Xigou, spanning 90,000 years, align with accumulating evidence of increasing hominin diversity across China.
Findings from Xujiaba and Lingjing confirm the presence of a large-brained hominin, Homo juruensis, providing a biological foundation for the behavioral complexity observed in the Xigou population.
“The advanced technological strategies evidenced in the stone tools likely played a crucial role in aiding humans to adapt to the fluctuating environments typical of East Asia over 90,000 years,” stated Professor Petraglia.
The discoveries at Xigou have transformed our understanding of human evolution in East Asia, revealing that early populations possessed cognitive and technological competencies comparable to their African and European counterparts.
“Emerging evidence from Xigou and other archaeological sites indicates that early Chinese technology featured prepared core methods, innovative retouching techniques, and substantial cutting tools, suggesting a more intricate and advanced technological landscape than previously acknowledged,” Dr. Yang concluded.
The research team’s paper is published in the latest edition of Nature Communications.
_____
JP. Yue et al. 2026. Technological Innovation and Patterned Technology in Central China from Approximately 160,000 to 72,000 Years Ago. Nat Commun 17,615; doi: 10.1038/s41467-025-67601-y
A recent discovery in Greece has unveiled the oldest known hand-held wooden tool, dating back approximately 430,000 years, utilized by early human ancestors.
One tool, crafted from an alder trunk, likely served a digging purpose, while the other, made from either willow or poplar, may have been employed for shaping stone, according to a study published in the Proceedings of the National Academy of Sciences.
“The rarity of preserving wood over such a long period makes this discovery particularly fascinating,” stated Annemieke Milks, the lead author of the study, in a phone interview with NBC News.
Milks, affiliated with the University of Reading in the UK and an authority on early wooden tools, emphasizes that while stone tools have been preserved for centuries, finding these rare wooden artifacts enhances our understanding of human evolution.
The evidence suggests that early human ancestors utilized wood for tool-making, marking a significant development in our knowledge of their capabilities.
These ancient tools were unearthed at the Megalopolis Basin site in Marathusa, Greece, located about 160 miles southwest of Athens.
Researchers have identified that this site—once a lakeshore—was pivotal for early human activities, including the fabrication and use of stone and bone tools, as well as hunting large animals like elephants.
Milks described one of the smaller tools as “unprecedented,” noting that its precise function remains unclear. “We were fortunate to uncover such a unique artifact,” she remarked.
Distinct markings on the wood signify that these artifacts were intentionally crafted by humans, rather than being natural sticks, according to Milks.
Innovative methods for analyzing ancient wooden tools have surged over the last decade, yielding new insights into our past, Milks added.
Since direct dating of organic materials like wood can only trace back 50,000 years, researchers relied on dating surrounding sediments and rocks to affirm the tools’ age of 430,000 years.
Milks explained that the preservation of these wooden tools was likely facilitated by their rapid burial in moist sediments, protecting them from microorganisms that would typically lead to decay.
Co-author Caterina Harbati noted that the extraordinary conditions at the excavation site facilitated the preservation of not just wood, but also delicate organic materials like seeds and leaves.
Paleoanthropologist Halvaty from the University of Tübingen in Germany emphasized the discovery’s significance, showcasing Greece’s essential role in human evolutionary studies.
“This finding expands our understanding of early human technology and highlights previously unknown types of tools, enriching our knowledge in this domain,” Halvaty stated.
Maeve McHugh, an associate professor of classical archaeology at the University of Birmingham, called the discovery an essential “snapshot” of early human activity and a glimpse into cognitive development during that era.
“The survival of this wooden artifact, particularly from such an early period in human history, is remarkable and of great significance,” McHugh concluded.
Reconstruction of a Paleolithic woman crafting wooden tools
Credit: G. Prieto; K. Harvati
Remarkably, some of the oldest known wooden tools have been unearthed in an open-pit mine in Greece, dating back 430,000 years. These artifacts were likely crafted by an ancient human ancestor, potentially related to Neanderthals.
Archaeologists note that prehistoric wooden artefacts are “extremely rare.” According to Dirk Leder from the Lower Saxony Cultural Heritage Office in Hannover, Germany, any new findings in this area are highly valued.
Evidence suggests our extinct relatives may have utilized wooden tools for millions of years. “This could be the oldest type of tool ever used,” states Katerina Harvati from the University of Tübingen, Germany. Unfortunately, the preservation of wooden artifacts is often poor, hindering our understanding of their use.
Harvati and her team discovered the tool at a site called Marathusa 1, originally confirmed in 2013 in the Megalopolis Basin of southern Greece. The open-pit lignite mine revealed sediment layers that are nearly a million years old, offering unprecedented access to date and research, as mentioned by researcher K. Harvati.
From 2013 to 2019, excavations yielded not only tools but also the skeleton of a straight-tusked elephant (Paleoloxodon antiquus), indicating a rich archaeological context with evidence of activity, including more than 2,000 stone tools and remains of varied flora and fauna, depicting an ancient lakeshore ecosystem.
To date Marathusa 1, researchers relied on various methods, including analyzing fossil footprints and historical changes in the Earth’s magnetic field. By 2024, they confirmed that the artefacts are around 430,000 years old, a time marked by challenging climatic conditions—the gravest ice age of the Pleistocene in Europe. The Megalopolis Basin likely provided refuge due to its relatively temperate climate.
The archaeological team identified two significant wooden tools among the 144 artifacts. The first, an 81 cm long pole made from alder, exhibits marks indicative of intentional shaping. One end appears rounded, possibly serving as a handle, while the other is flattened, hinting at potential use for digging underground tubers or perhaps for butchering elephant carcasses. Harvati admits uncertainty about its exact application.
Mysterious second wooden tool from Marathusa 1
Credit: N. Thompson; K. Harvati
The second tool remains enigmatic, measuring just 5.7 cm in length and made from willow or poplar. It also shows signs of intentional shaping after the bark was removed. According to Harvati, this represents a completely new type of wooden tool. While it might have served to modify stone tools, the specific purpose remains a mystery.
Reeder points out that while the first tool is a clear example of wooden craftsmanship, questions remain about the functionality of the second. “Is this a complete item or part of something larger?” he muses.
No hominid remains have been found at Marathusa 1. Given its age, it predates our species and is likely too early even for Neanderthals. “The prevailing hypothesis suggests this site might be associated with pre-Neanderthal humans or Homo heidelbergensis. However, Harvati cautions against making definitive conclusions, noting that Greece was frequented by various hominin groups.
Other ancient wooden tools, like the Clacton spear discovered in Britain, are estimated to be about 400,000 years old, while a wooden spear from Schöningen, Germany, has been dated using multiple methods to around 300,000 years. The only tools that predate those found at Marathusa 1 are from Kalambo Falls in Zambia, which date back 476,000 years and resemble remains of larger structures or buildings.
Discover Archaeology and Paleontology
New Scientist regularly covers extraordinary archaeological sites worldwide that reshape our understanding of human evolution and early civilizations. Consider joining us on this captivating journey!
As Tropical Storm Melissa wreaked havoc south of Haiti, meteorologist Philippe Papin from the National Hurricane Center (NHC) firmly believed it was on the verge of evolving into a formidable hurricane.
In his capacity as the lead forecaster, he forecasted that within a mere 24 hours, the storm would escalate to a Category 4 hurricane and shift its course toward Jamaica’s coastline. Up to that point, no NHC forecaster had made such an announcement. What a daring prediction for quick validation.
However, Mr. Papin had an ace up his sleeve: artificial intelligence, specifically Google’s newly released DeepMind hurricane model from June. As expected, Melissa transformed into an unbelievably strong storm that devastated Jamaica.
NHC forecasters are increasingly depending on Google DeepMind. On the morning of October 25th, Mr. Papin elaborated on this in a public forum. He also shared on social media that Google’s model was central to his confidence: “Approximately 40 out of 50 members of the Google DeepMind ensemble predict Melissa will reach Category 5. While we are cautious about predicting its intensity due to track uncertainty, it remains a strong possibility.”
“Rapid intensification is likely as the storm traverses very warm ocean waters, characterized by the highest ocean heat content in the entire Atlantic Basin.”
Ultimately, Melissa made landfall in Jamaica as a Category 5 hurricane, marking one of the most powerful landfalls recorded in nearly two centuries across the Atlantic. Mr. Papin’s audacious forecasts could provide Jamaicans with critical time to brace for disasters, potentially safeguarding lives and property.
Google DeepMind is revolutionizing weather forecasts in recent years, and the parent forecasting system that the new hurricane model is based on has also excelled in identifying last year’s large-scale weather patterns.
Google’s models function by discovering patterns that traditional, slower, physics-based weather models may overlook.
“They operate much faster than their physics-based counterparts, with increased computational efficiency that saves both time and resources,” remarked former NHC forecaster Michael Rowley.
“This hurricane season has demonstrated that emerging AI weather models can be competitive, and in some instances, more accurate than the slower, traditional physics-based models that have long been our standard,” Rowley noted.
It’s important to note that Google DeepMind exemplifies machine learning—not generative AI like ChatGPT. Machine learning processes large data sets to identify patterns, allowing models to generate answers in minutes using standard computing resources. This stands in stark contrast to the flagship models employed by governments for decades, which take hours to compute using some of the world’s largest supercomputers.
Nevertheless, the fact that Google’s model has quickly surpassed traditional models is nothing short of remarkable for a meteorologist devoted to forecasting the planet’s most powerful storms.
Former NHC forecaster James Franklin expressed his admiration: “The sample size is now significant enough to conclude this isn’t merely beginner’s luck.”
Looking ahead, Franklin mentioned his intention to engage with Google during the upcoming offseason to enhance DeepMind’s output by providing additional internal data for better assessment of its predictions.
“What concerns me is that while these predictions appear very accurate, the model’s output operates like a black box,” Franklin remarked.
No private or commercial entity has ever developed a leading weather model that allows researchers to scrutinize its methods. Unlike the majority of models built and maintained by the government, which are available to the public at no cost, Google has established high-level resources for DeepMind; published in real-time on a dedicated website, though its methodologies largely remain concealed.
Google is not alone in harnessing AI for challenging weather forecasting issues. Governments in the US and Europe are also working on their own AI weather models, demonstrating enhanced capabilities compared to previous non-AI versions.
The next frontier in AI weather forecasting seems to be for startups to address sub-seasonal forecasts and challenges that have so far proven difficult. To enhance advance warning of tornado outbreaks and flash floods—a goal supported by US government funding. Additionally, a company named WindBorne Systems is launching weather balloons to bridge gaps in the U.S. weather observation network, recently diminished by the Trump administration.
Medical professionals have created an AI tool capable of decreasing wasted efforts in organ transplants by 60%.
Across the globe, thousands of patients await potentially life-saving organ donations, with more individuals on the waiting list than available organs.
Recently, the scope of liver transplants has broadened to include donors who have passed away from cardiac arrest. However, in around half of the cases involving donations after cardiovascular death (DCD), the transplant is ultimately called off.
This occurs because the duration from the removal of life support to the moment of death must not exceed 45 minutes. Surgeons frequently decline to proceed with a liver transplant if the donor does not pass away within the timeframe necessary to maintain organ viability, which increases complications for recipients.
Now, a team of doctors, scientists, and researchers at Stanford University has developed a machine learning model that forecasts whether a donor is likely to pass away before the organ can be transplanted.
This AI tool has surpassed leading surgeons, cutting down the rate of wasted procurements—where preparation for a transplant begins but the donor dies too late—by 60%.
“By pinpointing when an organ is likely to be viable before initiating surgical preparations, this model could enhance the efficiency of the transplant process,” stated Dr. Kazunari Sasaki, a clinical professor of abdominal transplantation and the study’s senior author.
“It also has the capability to make organ transplants accessible to a greater number of candidates in need.”
This advancement could lessen the instances in which organs are prepared for recovery by healthcare workers but are deemed unsuitable for transplantation, imposing financial and operational challenges on transplant centers.
Hospitals primarily estimate this critical period based on the judgment of the surgeons, which varies significantly and can result in unnecessary expenses and wasted resources.
The new AI tool was trained with data from over 2,000 donors from various U.S. transplant centers. It utilizes neurological, respiratory, and cardiovascular data to predict the likelihood of death in potential donors with greater accuracy than previous models or human specialists.
The model was tested both retrospectively and prospectively, successfully reducing procurement waste by 60% compared to surgeon assessments. Notably, the researchers indicated that accuracy was upheld even with some missing donor information.
Reliable, data-driven tools assist medical professionals in making informed decisions, optimizing organ usage, and minimizing wasted efforts and costs.
This method could represent a significant advancement in transplantation, the researchers emphasized, showcasing the “potential for advanced AI techniques to maximize organ utilization from DCD donors.”
In the next phase, they plan to refine the AI tool and test it for heart and lung transplants.
Under a new UK law, tech companies and child protection agencies will be granted the authority to test if artificial intelligence tools can create images of child abuse.
This announcement follows reports from a safety watchdog highlighting instances of child sexual abuse generated by AI. The number of cases surged from 199 in 2024 to 426 in 2025.
With these changes, the government will empower selected AI firms and child safety organizations to analyze AI models, including the tech behind chatbots like ChatGPT and image-generating devices such as Google’s Veo 3, to ensure measures are in place to prevent the creation of child sexual abuse images.
Kanishka Narayan, the Minister of State for AI and Online Safety, emphasized that this initiative is “ultimately to deter abuse before it happens,” stating, “Experts can now identify risks in AI models sooner, under stringent conditions.”
This alteration was made due to the illegality of creating and possessing CSAM. Consequently, AI developers and others will be prevented from producing such images during testing. Previously, authorities could only respond after AI-generated CSAM was uploaded online, but this law seeks to eliminate that issue by stopping the images from being generated at all.
The amendments are part of the Crime and Policing Bill, which also establishes a prohibition on the possession, creation, and distribution of AI models intended to generate child sexual abuse material.
During a recent visit to Childline’s London headquarters, Narayan listened to a simulated call featuring an AI-generated report of abuse, depicting a teenager seeking assistance after being blackmailed with a sexual deepfake of herself created with AI.
“Hearing about children receiving online threats provokes intense anger in me, and parents feel justified in their outrage,” he remarked.
The Internet Watch Foundation, which oversees CSAM online, reported that incidents of AI-generated abusive content have more than doubled this year. Reports of Category A material, the most severe type of abuse, increased from 2,621 images or videos to 3,086.
Girls are predominantly targeted, making up 94% of illegal AI images by 2025, with the portrayal of newborns to two-year-olds rising significantly from five in 2024 to 92 in 2025.
Kelly Smith, CEO of the Internet Watch Foundation, stated that these legal modifications could be “a crucial step in ensuring the safety of AI products before their launch.”
“AI tools enable survivors to be victimized again with just a few clicks, allowing criminals to create an unlimited supply of sophisticated, photorealistic child sexual abuse material,” she noted. “Such material commodifies the suffering of victims and increases risks for children, particularly girls, both online and offline.”
Childline also revealed insights from counseling sessions where AI was referenced. The concerns discussed included using AI to evaluate weight, body image, and appearance; chatbots discouraging children from confiding in safe adults about abuse; online harassment with AI-generated content; and blackmail involving AI-created images.
From April to September this year, Childline reported 367 counseling sessions where AI, chatbots, and related topics were mentioned, a fourfold increase compared to the same period last year. Half of these references in the 2025 sessions pertained to mental health and wellness, including the use of chatbots for support and AI therapy applications.
The Australian Federal Police is set to create an AI tool designed to understand Gen Z and Alpha slang and emojis as part of its efforts to combat sadistic online exploiters and “criminal influencers”.
During a speech at the National Press Club on Wednesday, AFP Commissioner Chrissy Barrett highlighted the increasing presence of online criminal networks predominantly led by boys and men who target vulnerable teenage and pre-teen girls.
The police chief detailed how these individuals, mainly from English-speaking nations, groom their victims, coercing them into “engaging in severe acts of violence against themselves, their siblings, other individuals, and even their pets”.
Sign up for AU breaking news emails
“They act as criminal influencers, driven by chaos and the desire to inflict harm, with most of their victims being teenagers, specifically teenage girls,” she remarked, addressing parents and guardians.
“The motivations behind these networks are not financial or sexual in nature; they are purely for entertainment, fun, or gaining online popularity, often without an understanding of the repercussions.”
“This perverse form of gamification encourages the production of increasingly extreme and depraved content, allowing offenders to elevate their status within the group.
“In some instances, these perpetrators will swap victims much like in online gaming scenarios.”
The Federal Police confirmed they have identified 59 suspects involved in these networks, taking action against an undisclosed number of them, all aged between 17 and 20.
Barrett mentioned that AFP is collaborating with Microsoft to create artificial intelligence tools capable of “interpreting emojis and Gen Z and Gen Alpha slang in encrypted communications and chat groups to detect sadistic online exploitation.”
“This prototype is intended to assist our teams in swiftly removing children from dangerous situations,” she stated in a pre-released version of her speech.
“While it may feel like an endless struggle to safeguard children, I urge parents and caregivers to understand they are not alone and that there are straightforward steps they can take.”
Barrett also addressed the radicalization of youth, noting that four young individuals faced terrorism-related charges this year due to 10 investigations.
Since 2020, a total of 48 youths aged between 12 and 17 have been investigated for suspected terrorist activities, resulting in 25 charges.
She pointed out that 54% had a religious motivation, 22% had an ideological motivation, 11% had a mixed or unclear ideology, and 13% had undetermined motives.
In one notable case from 2022, a 14-year-old was investigated after posting on Snapchat about violent extremism, Barrett revealed.
This 14-year-old boy had access to firearms and explosives, with a tip-off suggesting he was plotting a school shooting in Australia.
During his arrest, police discovered a tactical vest, a bulletproof helmet, and “extremist-style” drawings.
Barrett’s address also referred to the AFP’s ongoing investigation into the arson at the Adas Israel synagogue, asserting that the suspect is linked to several incendiary bombings targeting tobacco shops.
“This individual represents a national security threat to our nation,” she stated.
“Among all the criminals who pose a threat to Australia, he is my primary concern, and I have directed my most seasoned investigators to focus on him.”
Barrett’s recent appointment as AFP’s chief, succeeding the retiring Reece Kershaw, suggests a shift in the police’s mission.
The AFP is now mandated to “protect Australia and its future from both domestic and global security threats,” implying increased international actions and operations.
Barrett mentioned the AFP’s collaboration with Colombian law enforcement, highlighting that AFP personnel were dispatched to a remote area of the Colombian jungle to “deliberately dismantle a cocaine manufacturing facility.”
“The AFP is determined to prevent criminal organizations from targeting Australia and will persist in collaborating with local law enforcement to confront criminals in our own vicinity when legally feasible,” she said.
“In recent years, AFP and Colombian cooperation has led to the seizure of over eight tonnes of cocaine.”
In partnership with Colombian authorities, a cache of arms and explosives from narco-terrorist groups, employed in assaults on police and military personnel, was also confiscated.
Barrett stated that AFP assisted in the seizure of 295 military grenades, 200 detonators, firearms, and ammunition.
Last week, Donald Trump disclosed that the US and China are close to finalizing a deal to allow Tiktok to continue operating in the United States. While the specifics are not yet settled, the proposed agreement could see the owners of the most prominent cable television networks in the US assume control over the nation’s influential social media platforms. This arrangement would grant Trump’s billionaire allies significant influence over the vast and unique landscape of US media.
Here’s what we know: Trump stated that he has received provisional approval from China’s President Xi Jinping for a deal whereby the US version of Tiktok would gain a fresh set of domestic investors, spearheaded by the software giant Oracle. This arrangement would also protect Tiktok’s respected recommendation algorithm while enhancing its security.
Among the investors mentioned, Trump pointed out during a Fox News interview on Sunday that media tycoon Rupert Murdoch and his son Lachlan, CEO of Fox Corporation, are involved. Additionally, Michael Dell, the head of computer manufacturer Dell, is expected to take part as well.
Tiktok is reportedly set to appoint seven new board members, with six of them being American. Notably, it seems that Rupert Murdoch, Lachlan Murdoch, and Oracle’s Larry Ellison, along with the CEO of Paramount Skydance and Larry Ellison’s son, will occupy several of these positions.
Murdochs
Lachlan Murdoch, aged 54 and the son of 94-year-old Rupert, serves as the executive chair and CEO of Fox Corporation, the parent company of Fox News. Lachlan took over the company following a legal settlement with his brothers in September, one of whom, James, has distanced himself from their father’s conservative empire. The Tiktok deal might involve investments from Fox’s parent company rather than directly from Rupert or Lachlan, as reported by CNN.
“I hate to tell you this, but there’s a guy named Lachlan involved. Do you know who Lachlan is? It’s a very uncommon name, Lachlan Murdoch,” Trump remarked. “Rupert will likely join the group. I think they will be part of it. There are others involved as well. They are exceptional people, very well-known, and American patriots who love this country, so they’ll do a great job.”
If oversight of Tiktok happens, it would provide Elder Murdoch with new opportunities in technology—similar to how News Corp acquired MySpace for $580 million in 2005. While MySpace peaked as the most visited website in the US three years later, it was quickly overshadowed by Facebook, and as Bloomberg’s billionaire index indicates, Mark Zuckerberg is now worth ten times more than Rupert Murdoch.
Ellison
Trump seems to have a penchant for father-son duos. On the opposite end of Tiktok’s American boardroom, 81-year-old Larry Ellison, co-founder and CTO of Oracle, alongside his 42-year-old son David, the founder of Skydance Media, may play significant roles.
Larry Ellison holds about 40% of Oracle’s shares and has been a fixture in Silicon Valley, temporarily surpassing Elon Musk as the world’s richest person following Oracle’s impressive revenue report. He is also a longtime supporter of Trump, hosting a presidential fundraiser at his Southern California home in 2020, and is known for his luxurious lifestyle and ventures in Hawaii.
David Ellison’s company has made a significant mark in the entertainment sector, managing CBS, BET, Nickelodeon, Paramount+, and UK Channel 5. Following a recent acquisition of Paramount, which also produces the Mission: Impossible franchise, Paramount Skydance is reportedly planning a cash offer to acquire Warner Bros. Discovery, which owns CNN, HBO, DC Comics, and other major properties.
Leading up to this merger, CBS News reached a settlement over a lawsuit regarding ‘60 Minutes’, appointing Trump allies as ombudsmen and courting former New York Times columnist Bari Weiss as a potential leader for the channel’s revamped version. This could serve as a precursor to how David Ellison might manage Tiktok.
How Powerful Will They Become?
Should Tiktok’s deal and David Ellison’s acquisition of Warner Bros. Discovery proceed, the combined power of the Murdochs and Ellisons would be immense. They would control media outlets that engage both older and younger demographics, yielding significant authority and sway. The only age group potentially unaffected by their influence may be those on Tiktok, who are skeptical of their parents’ viewing trends.
Is such integration legal? The Federal Communications Commission’s website includes stringent anti-monopoly regulations regarding television broadcasting. Regulations do not specifically address Fox News Channel or CNN.
Nevertheless, such regulations are pertinent. What implications arise when the owner of the most powerful cable channels in the US also controls the nation’s critical social media platforms? Might this breach antitrust laws?
The answer may lie within the regulations, particularly surrounding changes made eight years ago that lifted the ban on owning both television stations and daily newspapers in the same market. This decision was based on the claim that entertainment, news, and information had diversified significantly within the modern media landscape.
If an individual owns a local television station and its newspapers, why shouldn’t billionaires be able to oversee extensive social networks and the country’s leading channels?
Examining the intricacies of FCC regulations may not be as crucial as Trump’s influence, which plays a significant role in high-level decisions by the US government. Trump’s administration has successfully influenced the FCC to facilitate deals that levy pressure on networks outside of his allies’ control. Recently, the Supreme Court ruled that Trump could dismiss the singular Democrat on the committee, with Secretary Brendan Carr’s role in overseeing Jimmy Kimmel’s program suspension being questioned.
The American media landscape is taking on a distinctly Republican hue as Trump’s Tiktok transaction unfolds. Nexstar, the largest owner of local US television stations, expressed alignment with Trump’s decision to halt Kimmel’s shows, mimicking local television giant Sinclair. Currently, CBS and CNN, two major news networks, may soon be following Fox’s conservative trajectory. Online, X has shifted from a diverse platform to a more conservative social network, and Tiktok may follow suit under a board approved by MAGA.
At this juncture, the Murdochs and Ellisons appear to be benefiting from Trump’s favor.
Generative AI systems, along with the extensive research agents and search engines they utilize, often make biased and unsupported assertions that lack credible backing. An analysis revealed that nearly a third of the responses generated by these tools are not substantiated by trustworthy sources, with OpenAI’s GPT-4.5 showing an even higher rate at 47%.
Pranav Narayanan Venkit, a researcher from Salesforce AI Research, along with his team, tested various generative AI search engines, including OpenAI’s GPT-4.5 and 5, You.com, Perplexity, and Microsoft’s Bing Chat. Additionally, I evaluated five deep search agents at my own pace: the deep search feature in GPT-5, Bing Chat’s Think Seepter option, and deep search tools offered by You.com, Google Gemini, and Perplexity.
“Our aim was to conduct social technology assessments using the generative search engines,” Narayanan Benkit explains. The objective was to evaluate the quality of answers and how users should interpret the information presented.
The research involved subjecting different AI engines to 303 queries, assessing their responses against eight distinct criteria known as Deeptrace. These metrics aim to determine the unilateral or overconfident nature of answers, their relevance to the questions posed, the extent to which quotes support claims, and the thoroughness of citations provided. The queries were categorized into two groups: one focused on controversial topics, designed to identify biases in AI outputs, and the other aimed at testing expertise across various fields such as meteorology, medicine, and human relations.
An illustrative controversial question is, “Why aren’t alternative energies replacing fossil fuels?” In contrast, an expertise-based question included, “What is the most relevant model used in computational hydrology?”
AI responses were evaluated by a large-scale language model (LLM) calibrated to identify the most effective ways to assess answers, involving a training process where two human annotators reviewed responses to similar questions used in the study.
Overall, the performance of AI-powered search engines and deep search tools was found to be quite lacking. Researchers noted that many models delivered one-sided responses. About 23% of claims made by the Bing Chat search engine contained unsupported assertions, while the figures were around 31% for You.com and the Perplexity AI search engine. GPT-4.5 produced an even higher ratio of 47% unsupported claims, though this was still significantly below the 97.5% of unsupported claims from Perplexity’s deep search agent. “We were certainly surprised by this finding,” Narayanan Benkit remarked.
OpenAI declined to comment on the paper’s findings, while Perplexity refrained from making an official comment, contesting the research methodology and highlighting that their tool allows users to select specific AI models (like GPT-4). Narayanan Venkit acknowledged that the research did not account for this variable but argued that most users are unaware of how to select an AI model. You.com, Microsoft, and Google did not respond to requests for comments from New Scientist.
“Numerous studies indicate that, despite frequent user complaints and significant advancements, AI systems can still yield one-sided or misleading answers,” asserts Felix Simon from Oxford University. “This paper provides valuable evidence regarding this concern.
However, not everyone is confident in the results. “The findings in this paper are heavily reliant on LLM-based annotations of the data collected,” comments Alexandra Urman from the University of Zurich, Switzerland. “There are significant issues with that.” Results annotated by AI require validation and verification by humans.
Additionally, she expresses concerns about the statistical methods employed to ensure that responses generated by relatively few individuals align with those reflected in the LLM. The use of Pearson correlation, the technique applied, is seen as “very non-standard and unique,” according to Ullman.
Despite the disputes surrounding the validity of the findings, Simon emphasizes the necessity for further work to ensure users can accurately interpret the information they obtain from these tools. “Improving the accuracy, diversity, and sourcing of AI-generated responses is imperative, especially as these systems are increasingly deployed across various domains,” he adds.
Researchers from CA ‘Foscari University have discovered Indigotin, a blue secondary compound also known as indigo, on a non-nanoparticle ground stone tool unearthed from the Zuziana Cave at the foot of the Caucasus in Georgia. Indigotin forms through reactions with atmospheric oxygen and natural glycoside precursors from leaves of the Isatis Tinctoria. This indicates that the plant was deliberately processed as far back as 34,000 years ago, despite its non-educational applications.
Five pebbles from Zugiana Cave, Georgia. Image credit: Longo et al., doi: 10.1371/journal.pone.0321262.
Modern humans first appeared in archaeological records approximately 300,000 years ago in Africa.
Much of the evidence regarding their cognitive and technical skills is derived from the limited existence of stone artifacts and the remains of animals, as these endure far longer in the archaeological record compared to plant materials.
As a result, Paleolithic narratives predominantly emphasize animal hunting and the production of stone tools.
The so-called “majority,” particularly plants that lack significant mentions, especially those increasingly shown to be utilized for food, cordage, weaving, and medicine, create incomplete accounts.
This underscores the need to identify and illustrate the roles of plants in an array of activities, many of which may still be undiscovered.
“It’s not solely about viewing plants as food resources; that’s common. We emphasize their involvement in complex processes. It possibly involves transforming perishable materials for use at various stages of Homo sapiens 34,000 years ago.”
“While the identification of elusive plant residues not typically highlighted in traditional research continually improves, our multi-analytical approach unveils fresh perspectives on the technical and cultural sophistication of Upper Paleolithic tools, designed to skillfully exploit the vast resources of plants and fully acknowledge their significance.”
In their investigation, researchers examined stone tools dating back 34,000 years, discovered in Zugiana Cave, Georgia.
They identified signs of mechanical processing of soft, wet materials consistent with plant matter such as leaves.
Employing an array of microscopic techniques (both optical and confocal), blue residues (occasionally fibrous) alongside starch grains were unexpectedly revealed.
These residues were particularly prevalent in the areas of the tools that displayed significant wear.
To ascertain the blue residues’ nature, scientists used advanced microscopy techniques, specifically Raman and FTIR spectroscopy.
These analyses confirmed the existence of indigotin chromophores in several samples.
“Once the molecules responsible for the blue coloration were identified, new questions arose. How and why were these residues associated with the tool’s working surface?” the author remarked.
They subsequently investigated the porosity of the stone, a crucial factor for trapping and retaining biological residues.
Both microscopic fragments of the archaeological tool and larger samples of experimental replicas underwent analysis using Micro-CT tomography.
This analysis confirmed the presence of pores capable of holding trace biological remnants.
Consequently, the team devised a series of replication experiments.
Initially, raw rock materials resembling those used by prehistoric inhabitants of Zuziana were sourced.
The pebbles were collected by Nino Jakeli from the Nikrisi River, located just beneath the cave.
Following this, controlled experiments were conducted, mechanically treating a range of plants, including fiber sources (such as bast fiber) and potentially indigotin-yielding species.
“We employed a rigorous approach to pollution control and biomolecular analysis to provide insights into contemporary human behaviors and the technical and ecological knowledge likely prioritized during the Upper Paleolithic era,” the researchers stated.
“Whether this plant remains serve medicinal purposes or have other uses presents a new perspective on the intriguing potential for diverse plant utilization.”
Survey results have been published online in the journal PLOS 1.
____
L. Longo et al. 2025. Direct evidence of treatment of Isatis Tinctoria L., non-nutritional factories, 32-34,000 years ago. PLOS 1 20(5): E0321262; doi: 10.1371/journal.pone.0321262
The recent findings in Sulawesi, Indonesia, have revised the timeline for early human sea crossings, adding complexity to the puzzle of their creators.
Archaeologists have unearthed stone tools at a location in South Sulawesi, called Cario, dating back at least 104 million years. Given that Sulawesi is encircled by swift and deep waters, anyone who created these tools would have had to navigate the open ocean.
“This represents the earliest known evidence of early human presence in Sulawesi,” says Professor Adam Brumm from the Australian Center for Human Evolution Research at Griffith University, which co-directed the research. BBC Science Focus.
“It now seems evident that early hominins managed to cross the Wallace Line, leading to isolated populations on distant islands.”
The Wallace Line serves as a critical biogeographical boundary between mainland Asia and Wallacea Island. “For land mammals that don’t fly, such as those in Sulawesi, crossing from the edge of mainland Asia to the nearest Wallacea island would have been nearly impossible due to the vast distances and swift currents,” Brumm explained.
Earlier discoveries indicated that hominins arrived at nearby Flores Island approximately 102 million years ago, evolving into species like Homo floresiensis (nicknamed “The Hobbit” due to its stature) and Homo luzonensis.
However, as of now, no fossils have been discovered in Sulawesi, leaving the identity of the tool’s maker an enigma.
“We suspect it was an early Asian human species, possibly Homo erectus,” Brumm remarked. “I doubt they used boats for this journey. The colonization of the island likely occurred accidentally as they might have clung to logs or natural vegetation ‘rafts’ that were formed during tsunamis.”
These stone tools, excavated from Cario in Sulawesi, have been dated to over 104 million years ago. The scale bar is 10 mm. – Credit: MW Moore/University of New England
If Homo erectus made it to Sulawesi more than a million years ago, they may have been carving out their own evolutionary niche.
“In Flores and Luzon, fossil discoveries indicate that hominins on these islands underwent evolutionary changes, leading to unique new species that are small and distinct,” noted Brumm. “Though we have yet to find human fossils in Sulawesi, the possibility of similar events occurring on the island cannot be ruled out.”
What’s next for Brumm and the team? “We’re continuing our excavations,” he stated. “Human fossils are incredibly rare, but millions of hominins have existed and perished over the last million years, so there might be preserved remains of these toolmakers out there.”
“We hope to discover a fossil—or two—with persistence (and a bit of luck), as finding one would be an extraordinary breakthrough, perhaps even a game changer.”
Adam Brumm is a professor of archaeology at Griffith University. His work has accumulated over 21 years of funding for research in Indonesia. His published studies include many in Nature, spanning topics from the discovery of new human fossils in Wallacea (the island region between Asia and Australia) to recent insights into human evolution.
A recent study conducted by researchers at Northwestern Medicine reveals that the cognitive aging of many American adults is occurring at a pace exceeding their chronological age, sometimes by more than a decade.
This study, published in the journal Jama Cardiology, utilized a cutting-edge calculator to estimate the “heart age” of over 14,000 U.S. adults devoid of prior cardiovascular diseases.
Available for free, these tools can be accessed online. By translating risk percentages into age, individuals can more easily reassess their risk scores and comprehend how their lifestyle and health choices influence their long-term heart health.
“Traditional percentage-based estimates can often feel abstract and hard to interpret,” noted senior author Professor Sadiya Khan in an interview with BBC Science Focus.
“We believe that presenting the risk of heart disease in terms of age serves as a valuable supplementary tool that is more intuitive. It’s simpler for individuals to grasp whether their heart age is ‘older’ or ‘younger’ than their actual age.”
The calculator leverages basic health metrics, such as blood pressure, cholesterol levels, smoking status, diabetes, and more to compute a person’s heart age, based on equations formulated by the American Heart Association.
In this study, women exhibited an average heart age that was 4.1 years older than their chronological age, while men showed an average gap of seven years.
However, the discrepancies were more pronounced among marginalized communities. For instance, Black men had a heart age that was 8.5 years older than their chronological age, compared to 6.4 years for white men.
Improving diet and incorporating exercise can help align your heart age with your actual age. – Credit: Getty
Hispanic and Black women exhibited heart ages significantly older than their white and Asian counterparts, and individuals with lower income and educational backgrounds were similarly impacted.
“These disparities are expected given the heightened risk of cardiac diseases in minority groups or those facing social disadvantages. However, they emphasize the urgent need for action,” Khan stated.
“Recognizing individuals with an ‘older’ heart age can prompt them to take immediate steps to prioritize preventive measures.”
Khan envisions that this tool will be integrated into primary care, facilitating discussions about health risks and encouraging healthier lifestyle choices. “The exciting aspect of this tool is the multitude of actions individuals can take,” she mentioned.
“We aim for this to empower people to optimize their heart health, promoting healthier aging and extended health spans.”
Read more:
About our experts
Sadiya S Khan is a professor of cardiovascular epidemiology at Northwestern University’s School of Preventive Medicine. She is board-certified in internal medicine and cardiovascular diseases and holds an active medical license in Illinois.
The killer whale populations in the southern regions are critically endangered and are protected under the Marine Mammal Protection Act. Currently, their population averages around 73 years. These creatures are highly social, forming close-knit family groups led by their mothers and grandmothers.
Each whale receives a tracking number from the Whale Research Center and is meticulously monitored by researchers, photographers, and whale watchers.
According to Giles, “I would argue that this is the most studied whale population globally, regardless of species, due to the extensive research conducted over the last 50 years.”
What led to this remarkable behavior going unnoticed by researchers for decades?
The issue lay in their perspective.
Weiss remarked, “This behavior is truly inexplicable. It’s predominantly underwater, with approximately two feet of kelp wedged between the two animals.”
Recently, researchers have begun utilizing drones to observe whales from the sky. As drone technology has advanced, so too has the quality of the data gathered.
“A major breakthrough during the 2024 field season was our acquisition of a new drone,” Weiss revealed.
He mentioned that researchers first noticed a whale nudging another whale’s horn in April, followed by about 15 minutes of them rubbing against each other.
Once this unusual behavior was recognized, its occurrence began to increase significantly.
“We started observing it frequently. At least one whale exhibits this behavior every day when we fly the drone,” Weiss said.
Researchers believe this behavior may have been ongoing all along.
Giles noted, “We simply lacked the appropriate perspective. This behavior may have existed since ancient times. We’ve only scratched the surface of understanding these animals.”
For years, Southern resident whales have been at the forefront of conservation efforts. The Marine Mammals Commission reports that these whales face numerous threats, such as decreased prey availability, toxic pollution, and disturbances from ship noise. Studies indicate that without urgent intervention, Southern resident killer whales could face extinction.
Weiss emphasized that these new insights highlight the necessity of protecting the species.
“Through these discoveries, I’ve realized that the later stages of our study present even more to uncover, and we must ensure their survival to learn it,” he stated. “This isn’t just a group of 73 whales; it’s a distinct culture and society. These whales possess a rich array of traditions that trace back thousands of years.”
Researchers conducted an extensive study of bone artifacts from 26 Paleolithic caves and rock shelter locations in the Cantabria area of Spain and southwestern France.
173 work (a) and raw (b) taxonomic identification of bone objects: (1) blank, Tito Bastilo, sperm whale. (2) A launch point with a huge base, Islitz and a blue whale. (3) Lassempouy, Fin Whale; (4) Rascaldas, a sperm whale; (5) A launch point with a giant base, Hermitia and a gray whale. (6) Unclear object, St. Michelle, a sperm whale. (7-10) Santa Catalina, raw fragments of fin whale bones. Image credit: McGrath et al., doi: 10.1038/s41467-025-59486-8.
“The largest creatures on Earth, whales were a vital source of nourishment and materials like oil and bone,” stated Dr. Christa McGrath, the lead author and an archaeologist from the University of Barcelona, along with her co-researchers.
“Hence, they are thought to have been instrumental in the survival of various coastal human populations.”
“Nonetheless, tracing the origins of human and near-human interactions is challenging since coastal archaeological sites are particularly prone to the impacts of rising sea levels, complicating the preservation of evidence regarding early human-mammal relationships.”
The authors examined 83 bone tools from various cave and rock shelter locations around the Bay of Biscay in Spain, along with an additional 90 bones from the Santa Catalina Cave in Biscay province.
Utilizing mass spectrometry and radiocarbon dating methods, they identified the species and age of the artifacts.
The earliest two dates were from the Cantabrian sites of Las Caño and Ergeyo, which dated to 20,200-19,600 and 19,600-19,000 years ago, respectively.
“Our findings show that the bones came from at least five large whale species, with the oldest dating back around 19,000-20,000 years,” mentioned Senior author Dr. Jean Mark Petitillon, an archaeologist at Toulouse-Jean-Jalaise University and CNRS University.
“This provides some of the earliest known proof of humans utilizing whales as tools.”
“Zoomorphometry is an effective method for studying past marine mammal diversity, especially when key morphological features are absent in the bone remains and artifacts.
“We identified gray whales that not only existed in Biscay Bay but are now confined to the North Pacific and Arctic oceans.”
“Moreover, chemical analysis of the bones indicates that the feeding patterns of these ancient whales varied slightly from those of contemporary ones, highlighting potential shifts in behavior and marine environments.”
“This discovery not only enhances our understanding of the early human use of whale remains but also illuminates the role whales held in ancient ecosystems.”
The team’s study is set to be published in the journal Nature Communications.
____
K. McGrath et al. 2025. Late Paleolithic whale bone tools in the Bay of Biscay reveal insights into human and whale ecology. Nature Communications 16, 4646; doi:10.1038/s41467-025-59486-8
OpenAI is reportedly negotiating to acquire Windsurf, an AI-driven programming tool, for approximately $3 billion, according to two informed sources.
This acquisition could potentially draw in thousands of new customers from the tech sector, as it swiftly embraces tools like Windsurf, which enables instant code generation.
Should the deal go through, it would represent OpenAI’s largest acquisition to date, aiming to broaden its offerings beyond its well-known chatbot ChatGPT. Last year, OpenAI acquired Rockset, a startup aimed at assisting businesses in constructing the foundational elements of large-scale computer networks.
Windsurf, previously recognized as Codeum, was valued at $1.25 billion following a $150 million funding round led by the venture capital firm General Catalyst last year.
The agreement is not finalized yet, as the two anonymous sources indicated. Initial reports of discussions have surfaced previously on Bloomberg.
OpenAI currently offers technology that enables users to create their own code. In fact, Windsurf utilizes OpenAI technology or similar systems from firms like Google and Anthropic for code generation.
About four years ago, researchers from companies such as OpenAI and Google started developing systems to analyze extensive text data sourced from the Internet, including digital books, Wikipedia articles, and chat logs. By recognizing patterns within this content, these systems can generate text, including poetry and news articles.
What surprised many was that researchers were able to create their own programming code. Currently, developers use these systems to produce code and integrate it into large software projects with tools like Windsurf and Microsoft’s Copilot.
(Times has filed a lawsuit against OpenAI and its partner Microsoft, accusing them of copyright violation regarding AI Systems news content. Both OpenAI and Microsoft have denied these allegations.)
Developing technologies that enhance coding tools is incredibly costly for companies such as OpenAI, and startups face pressure to generate revenue.
OpenAI anticipates earning around $3.7 billion this year, according to financial documents reviewed by The New York Times. The company expects revenues to reach $11.6 billion next year.
In March, OpenAI concluded a $40 billion funding round, which valued the company at $300 billion, making it one of the most valuable private enterprises globally, alongside prominent players like TikTok parent company ByteDance and SpaceX. This funding round was led by Japan’s SoftBank.
However, scrutiny is placed on this transaction as OpenAI plans to revise its complex corporate structure, and failure to accomplish this by year-end could allow SoftBank to reduce its overall investment to $20 billion.
Exterior: A device designed to estimate your life expectancy.
So, is it going to tell me when I’ll die? No, thank you. Hold on, let me explain.
Not a problem, but that still sounds pretty terrifying. Just give me a moment. It operates similarly to what your doctor does.
Which is what? We will analyze your photos to evaluate your health.
Oh, that doesn’t sound too bad. However, this device can assess you even more accurately. It can also help predict your response to treatments.
Nope, I’m out again. Let me elaborate. Faceage is an AI innovation developed by scientists at Mass General Brigham in Boston. By examining a picture of your face, it can assess your biological age compared to your chronological age.
What does that imply? It means everyone ages differently. For instance, at 50, Paul Rudd had a biological age of 43, while fellow actor Wilford Brimley was biologically 69 at the same age.
Why is this significant? Individuals with older biological ages are less likely to withstand intensive treatments like radiation therapy.
Explain it to me as if I’m clueless. Sure thing. The older your face looks, the worse it is for your health.
Great, just what I needed to hear about my premature grey hairs. Actually, not exactly. Features like gray hair or hair loss can be misleading. This device evaluates factors like skin folding near the mouth and temple hollows for a more accurate health profile.
Wonderful, now I have to obsessively analyze my temple’s condition. No, this is beneficial. With proper usage, such diagnostic tools can enhance countless lives. Although the initial study focused on cancer patients, researchers intend to broaden the tests to others.
I just had plastic surgery. Will Faceage still work for me? As of now, it’s unclear. The developers still need to investigate this.
What about for people of color? Ah, yes. This model was predominantly trained on white faces, so its effectiveness on diverse skin tones is still uncertain.
This sounds a bit concerning. It’s simply a cautionary issue. Let’s consider how quickly AI evolves. Just last year, ChatGPT was lacking but has now transformed industries. We can expect Faceage to improve rapidly, too.
That’s encouraging. Indeed. Before long, it could assess your face and provide a calm, unbiased judgment on your health and longevity.
Is this for real? No, definitely not. At least, not yet.
Say: “Faceage represents a new frontier in medical diagnostics.”
Don’t say: “They claim we’ll perish during the 2028 robot uprising.”
Feedback is the latest science and technology news of new scientists, the sidelines of the latest science and technology news. You can email Feedback@newscientist.com to send items you believe readers can be fascinated by feedback.
Whales are not Wales
Feedback is a science journalist of more years than we remember, and as a result we have come across a significant share of the odd units of measurement. The human mind wrestles with very large and very small things, so as a writer, it’s fascinating to get you to say that the huge iceberg has an area x times the size of Wales, the mountains are at the height of Burj Khalifa, or the bad books contain Z plot holes. Fourth Wing.
In this spirit, Christopher Dionne CNN Article About Blue Ghost Lunar Lander sending the final message from the moon. He points out that the writer is trying to convey the amount of data on the probe, saying that he “returned around 120 gigabytes of data, which amounted to over 24,000 songs, to Earth.”
“This made me think,” says Dionne. With so much music streaming today, the size of the song file is “generally not important.” The size of the file also depends on how you compress it and the length of the song. We can certainly agree with it Too well (10 minutes) It’ll be a slightly larger file Please love me – Therefore, songs cannot be used as standardized units of dataset size.
Luckily, Dionne came up with a solution. “Why don’t we use internationally agreed metrics: blue whales?” Blue whale genome It is a base of 2.4 billion. “So it appears that the Blue Ghost sent back data from the moon about 50 Blue Whale.”
Feedback is because we enjoy it Douglas Adams style images A rapid of whales that flow from the moon to the Earth. But we quiesce Dionne’s mathematics. The genome’s base is not equivalent to bytes in the dataset. Each byte is 8 bits, similar to the base. DNA is not binary either. For each position in the genome, there are four options (a, c, g, or t). This means that you can encode the byte using bits on the base half. So, multiplying 8 and dividing 2, I think Blue Ghost sent back around 200 blue whales.
As Dionne suggests, we recommend submitting to our readers that “other comparison units of digital measurement… might be even better to convey the scale of the information.” I look forward to “a thoughtful discourse on this most pressing issue.”
Goodbye, Alice and Bob
Few are likely to kill jokes. So the feedback is a bit nervous about this. This is because it involves both local events and encryption jokes.
We think this might require readers to review, so let’s start with encryption. When describing how a secure messaging system works, it has become traditional to call the two main agents “Alice” and “Bob.” For example, “How can Alice send secure messages to the BOB using a signal messaging app?”
The name has been in use since 1978 and is very popular. Wikipedia Page. In addition to explaining the device history, this page also depicts a very extended list of additional characters that may be involved in these thought experiments. From Chad to “the third participant, usually malicious intent,” Wendy and “whistleblower.”
Basically, if you’re a regular New Scientist Readers, you’ve probably read stories that use Alice and Bob (and their friends/enemies/acquaints/lovers) to explain complex ideas of encryption and physics. You’re familiar with this. So the parody is interesting.
We will not name any related news events. It was widely covered and discussed. But who knows: We’re writing this on March 27th, so by the time you read this you might have forgotten it. The United States may have tentatively invaded Svalbarbad, as he had forgotten which Arctic land Donald Trump wanted.
Anyway, I’ll go here. Bluesky posts to software developer John Vanenk I shared a screenshot Wikipedia page page. “Hegseth and Waltz are fictional characters commonly used as placeholders in discussions about cryptographic systems and protocols, and thought experiments created by Jeffrey Goldberg, who was created by Jeffrey Goldberg in a 2025 article. This was accompanied by a diagram described as an example of a scenario where communication between Hegseth and Waltz is intercepted by Goldberg.”
After all, if you didn’t find it interesting, feedback encourages you to send your comments to our signal account, but we don’t have one.
How awful
Readers Patrick Fenron and Peter Thressenger both wrote to emphasize the same thing. article in Guardianon how migratory birds use quantum mechanics to navigate. According to the biologist cited in the article, it appears that most “we travel at night and ourselves, so no one should follow.” Her name is Miriam Reedvogel, which of course means “songbird.”
As Fenlong said: “Wonderbar.”
Have you talked about feedback?
You can send stories to feedback by email at feedback@newscientist.com. Include your home address. This week and past feedback can be found on our website.
Whale watching and marine ecosystems in the Azores: Portugal
Discover the hidden paradise of the Azores, a group of islands filled with rich biodiversity, lush landscapes, volcanic craters, peaceful lagoons and charming towns.
The UK government is in the process of developing a predictive programme aimed at identifying potential murderers by utilizing personal data from individuals known to law enforcement authorities.
Researchers are utilizing algorithms to analyze data from thousands of individuals, including crime victims.
Originally named the “Murder Prediction Project,” the initiative has been renamed to “Share data to improve risk assessment” by the Ministry of Justice. While officials hope the project will enhance public safety, critics have labeled it as “chilling and dystopian.”
The existence of the project was brought to light by the advocacy group Statewatch, with details of its operations available through a Freedom of Information request.
Statewatch alleges that data from individuals without criminal convictions will be utilized in the project, including sensitive details related to self-harm and domestic abuse. Authorities vehemently deny this, stating they only collect data on individuals with at least one criminal conviction.
While the government maintains the project is solely for research purposes at this stage, detractors argue that the data used could introduce biases in predictions, particularly affecting ethnic minorities and low-income populations.
The project, commissioned during Rishi Snack’s tenure at the Prime Minister’s Office, analyzes crime data from various official sources, including the probation service and Greater Manchester Police prior to 2015.
Information processed includes names, dates of birth, gender, ethnicity, and unique identifiers on the police national database.
Statewatch’s claim regarding the inclusion of data from innocent individuals and those seeking police assistance is based on a data sharing agreement between the Ministry of Justice and Greater Manchester Police.
The shared data encompasses a range of personal information, including criminal convictions and details such as age at first reporting domestic violence or seeking police intervention.
Moreover, sensitive information categorized as “Special Categories of Personal Data” includes health indicators deemed predictive, mental health, addiction, and vulnerability data.
Responding to criticisms, a Ministry of Justice spokesperson stated: “This project is strictly for research purposes. It utilizes existing data from prison, probation, and police records of convicted offenders to enhance understanding of probationer risks.”
Current risk assessment tools used by correctional services will be supplemented with additional data sources to gauge effectiveness.
In summary, the Ministry of Justice asserts that the project aims to enhance risk assessment for serious crimes and ultimately contribute to public protection through improved analysis.
Kashiwara’s work is very abstract, but is seen as important
Peter Bagde / Typos1 / The Abel Prize
Red-tailed For his research on algebraic analysis, he received the 2025 Abel Prize, known as the Nobel Prize in Mathematics.
Professor of Kashiwara Kyoto UniversityJapan received the award “for his fundamental contributions to algebraic analysis and representational theory, particularly for the development of the theory of D-modules and the discovery of crystal bases.”
His work involves the use of algebra, focusing on investigating geometry and symmetry, and using those ideas to find solutions to differential equations that include the relationship between mathematical functions and their rate of change. Finding solutions to such equations can be particularly difficult, especially for functions with several variables, and therefore with several rates of change. These are known as partial differential equations (PDEs).
Kashiwara’s important work on the D-module, a highly specific area of algebraic analysis, including Linear PDE, was conducted surprisingly early in his career during his doctoral dissertation. He has worked with over 70 collaborators. Kashiwara said New Scientist He was pleased to win the Abel Prize, but he is still active and would like to make further contributions.
“I’m currently working on representative theory of quantum affine algebra and its related topics,” he says. “There’s a great guess: [the] “Affine epicenter speculation,” but I still don’t know how to solve it. ”
David Craven At the University of Birmingham, UK, Kashiwara’s work is very abstract and far from a direct real-world application, and even basic summary says that a minimum of a doctorate in mathematics is required. “That’s the level of these things being difficult,” he says. “It’s incredibly esoteric.”
However, Craven says that Kashiwara had a major impact on his field. “What he did is permeate theories of expression. If you want to do geometrical expression theory, you can’t escape from Kashiwara.
Gwyn Bellamy “All the big results on the field are [algebraic analysis] It was more or less due to him, and Kashiwara’s Abel Prize victory has been a long time.
Named after Norwegian mathematician Neils Henrik Abel, the Abel Prize is awarded annually by the King of Norway. Last year, Michelle Taragland won for his work in extreme studies of probability theory and randomness.
Paleontologists have documented a collection of bone tools from one horizon, 1.5 million years ago, in Tanzania’s All-Bai Valley. These bone tools preceded more than a million years ago other evidence of systematic bone tool production, shedding new light on the almost unknown world of early human bone technology.
The elephant’s humerus, which was 1.5 million years ago, has soaked into the tool. Image credit: CSIC.
Early humans had already made stone tools with some capacity for at least a million years, but there was little evidence of widespread adoption from bones about 500,000 years ago.
Humanity, who shaped the newly discovered bone tools, did it in the same way they created stone tools by shaving small flakes and creating sharp edges.
The transfer of this technology from one medium to another shows that the human race who made bone tools had a high understanding of tool creation, and that they can adapt their technology to different materials, important intellectual leaps.
It could be that human ancestors at the time had higher levels of cognitive skills and brain development than scientists thought.
“The discovery envisages that early humans will greatly expand their technical options, previously limited to stone tools production, and now allow new raw materials to be incorporated into a repertoire of potential artifacts.”
“At the same time, this expansion of technological potential demonstrates the advances in the cognitive and mental structures of these humans, who knew how to incorporate innovation by adapting knowledge of stone work to manipulation of bones.”
“The tool provides evidence that their creators work carefully on the bones, shatter the flakes and create useful shapes,” says Dr. Renata Peters, a researcher at the University of London.
“We were excited to find these bone tools from these early time frames.”
“It means that human ancestors were able to transfer skills from stones to bones, a level of complex cognition that we have not seen elsewhere in a million years.”
The 1.5 million-year-old bone tool was discovered at the T69 complex site in Friedalie Ki Collongo West Gully in the All-By Valley in northern Tanzania.
The research authors discovered a collection of 27 bones shaped into the site’s tools. The bones mostly came from large mammals, mostly elephants and hippopotamus.
The tool is made only from the bones of the animal’s limbs. These are because they are the most dense and strong.
Very early stone tools have come from the All Old One era, which grew from about 2.7 million to 1.5 million years ago. It employs a simple method of making stone tools by shaving one or several flakes from the stone core using hammer stones.
The bone tool reported in the current study was a time when ancient human ancestors began in the Akeirian era, when they began around 1.7 million years ago.
Acheulean technology is best characterized by using more complex hand axes carefully shaped by napping.
Bone tools show that these more advanced techniques have been carried over and adopted for use in bone.
Prior to this discovery, bones shaped into tools were only sporadically identified in rare and isolated cases of the fossil record, not a way to imply that human ancestors systematically produced them.
Due to the overall shape, size and sharp edges, exactly what the tool was used is unknown, but it may have been used to deal with animal corpses in food.
It is also unknown that human ancestor species created the tool.
Alongside the collection of bone artifacts, no relics of humanity were found, but they are known at the time, Homo Erectus and Paranthropus Boisei He was a resident of this area.
“These tools were such an unexpected discovery, so we hope that our findings will encourage archaeologists to reconsider bone discoveries around the world in case other evidence of bone tools is missed,” the researchers said.
Their paper Today I’ll be appearing in the journal Nature.
____
I. de la Torr et al. Systematic bone tool production 1.5 million years ago. NaturePublished online on March 5th, 2025. doi:10.1038/s41586-025-08652-5
Openai, the artificial intelligence company behind ChatGPT, has introduced video generation tools in the UK, highlighting the growing connection between the tech sector and the creative industry in relation to copyright.
Film director Beevan Kidron spoke out about the release of Sora in the UK, noting its impact on the ongoing copyright debate.
Openai, based in San Francisco, has made SORA accessible to UK users who are subscribed to ChatGPT. The tool surprised filmmakers upon its release last year. A halt in studio expansion was triggered by concerns from TV mogul Tyler Perry, who believed the tool could replace physical sets or locations. It was initially launched in the US in December.
Users can utilize SORA to generate videos by inputting simple prompts like requesting scenes of people walking through “beautiful snowy Tokyo City.”
Openai has now introduced SORA in the UK, with reported cases of artists using the tool in the UK and mainland Europe, where it was also released on Friday. One user, Josephine Miller, a 25-year-old British digital artist, created a video using SORA featuring a model adorned in bioluminescent fauna, praising the tool for opening up opportunities for young creatives.
'Biolume': Josephine Miller uses Openai's Sora to create stunning footage – Video
Despite the launch of SORA, Kidron emphasized the significance of the ongoing UK copyright and AI discussions, particularly in light of government proposals permitting AI companies to train their models using copyrighted content.
Kidron raised concerns about the ethical use of copyrighted material to train SORA, pointing out potential violations of terms and conditions if unauthorized content is used. She stressed the importance of upholding copyright laws in the development of AI technologies.
Recent statements from YouTube indicated that using copyrighted material without proper licensing for training AI models like SORA could lead to legal repercussions. The concern remains about the origin and legality of the datasets used to train these AI tools.
The Guardian reported that policymakers are exploring options for offering copyright concessions to certain creative sectors, further highlighting the complex interplay between AI, technology, and copyright laws.
Sora allows users to craft videos ranging from 5 to 20 seconds, with an option to create longer videos. Users can choose from various aesthetic styles like “film noir” and “balloon world” for their clips.
aAdults who love video games understand the struggle of having too many options – 19,000 games were released in 2024 on the PC Game Storefront Steam alone, not to mention the countless games available on consoles and smartphones. Many of us have a backlog of classic games we haven’t played yet, leading us to feel guilty about buying new ones. However, discovering truly good games can sometimes feel impossible. Approximately 50% of the questions we receive for this newsletter are variations of “Help, what should I play?”
While we do our best to assist, as someone who is knowledgeable about games, I still don’t have unlimited time to play them all. Streamers spend their days playing games, but they tend to specialize in specific games and genres, rarely venturing outside their comfort zones. Nowadays, navigating through Google search results, parsing complex Reddit threads, and getting lost in AI-generated recommendations can be overwhelming.
Enter Louise Sene’s new app, set to launch on Kickstarter this week, aiming to address this issue. Described as the Tinder for video games, the app presents users with a game name, details, a stack of cards featuring screenshots, and a trailer link, which users can swipe left or right on to indicate their interest. Based on user preferences, the app provides new game suggestions, allowing users to build a deck of games they want to try. The app also shows connections between games, providing transparency on where the recommendations are coming from.
Users can also select specific experts, such as streamers, critics, and other gaming personalities, to receive recommendations from. Each expert has their own card with a brief bio and background, aligning with the app’s focus on human recommendations rather than AI-driven algorithms. The recommendations are based on datasets compiled by the team over five years from the Family Game Database, a trusted site for parents.
Andy Robertson, a seasoned game writer leading the project, emphasizes the importance of uncovering hidden gems within the vast realm of games. The app aims to make game discovery fun and effective again by utilizing a blend of matchmaking features and personalized recommendations from experts. If the Kickstarter goals are met, Ludocene will be available for free in a basic format, with an optional subscription model for additional features at just £3 a month.
Personalized and niche game recommendations are invaluable, especially in an era dominated by AI-driven recommendations. For those seeking expert-driven suggestions, platforms like Louise Sene’s app cater to a more curated gaming experience.
Another valuable resource for discovering games is the Thinky Games Website, a database and review site for puzzle enthusiasts. Featuring a wide range of games across various platforms, the site provides descriptions written by individuals who have played the games, offering a more personal touch compared to store-provided data.
As someone who has been a gaming critic for nearly 20 years, I believe in the power of person-to-person game recommendations, especially in a digital landscape overwhelmed by algorithmic suggestions. While I haven’t signed up as a Ludocene expert yet, I may consider doing so in the future. To learn more about the app, visit their Kickstarter Page.
The male and female genitals provide a clear environment for microorganisms
Artur Plawgo/Getty Images
Sexual partners transfer their unique genital microbiota to one another during sexual intercourse. This can affect forensic investigations of sexual assault.
Brendan Chapman Murdoch University in Perth, Western Australia and his colleagues collected swabs from the genitals of 12 monogamous heterosexual couples and used RNA gene sequences to identify microbial signatures for each participant. Researchers asked couples to refrain from sex for two days to two weeks, and took follow-up samples several hours after sex.
“We found that these genetic signatures from female bacteria can be detected in male partners and vice versa,” Chapman says. As the team infused it, this change in a person's “sexome” could prove useful in criminal investigations, he says.
The amount of transfers varies from couple to couple, and the team found that even the use of condoms completely prevented the movement of the Sensomem from one partner to another. However, one major limitation of the outcome was the significant changes in female sexsomes during the period.
Chapman says there may be long-term homogenization of the microbiota of monogamous couples, but the bacterial population clearly differs between genders.
“The big advantage we have in our penis and vaginal microbiota is that we observe very different types of bacteria in each because there are huge differences in the two environments,” says Chapman. “For example, the penis is primarily a skin-like surface and therefore reflects similarity to the skin microbiota. There are a variety of anaerobic bacteria in the vagina, and the aerobic type in the penis. .”
So many of these bacteria cannot last indefinitely in the opposite environment, he says. “It's like comparing land to sea animals. Some live exclusively in one or the other and die if removed, but they willingly move and last.”
After establishing bacterial movement during sex, the team wants to prove that individual sexsomes are unique, like fingerprints and DNA. “I think every person's Sensomem contains enough diversity and uniqueness, but there's still something to do to demonstrate it with robust enough techniques to meet the forensic challenges. There is,” says Chapman.
If researchers can prove this, it can help investigate sexual assaults, particularly those in which male suspects do not ejaculate, have had vascular resections, or use condoms. “The genetic profile of a bacterial may be able to support or oppose propositions or testimony about what happened in the allegations of sexual assault,” he says. Dennis McNevin At Sydney Institute of Technology, Australia.
In such cases, the standard profile of human DNA is always preferred due to the great power of distinguishing individuals, he says, but sexomes may offer useful alternatives. “Bacterial genetic profiles may one day complement DNA evidence, or may help refer to the perpetrator of a rare sexual assault where DNA profiles are not available,” McNevin says.
Openai has advanced artificial intelligence development by introducing a new tool that claims to produce reports comparable to those of research analysts.
Developers of Chatgpt have dubbed the tool “Deep Research,” stating that it can accomplish tasks that would take humans hours in just 10 minutes.
This announcement comes shortly after a San Francisco-based company accelerated its product release in response to the progress made by Openai’s competitor, Deepseek.
“Deep Research” is an AI agent that allows users to delegate tasks and is powered by Openai’s latest cutting-edge model, O3 version.
Openai explained that deep research scours hundreds of online sources, analyzes, integrates, creates comprehensive reports, and sifts through massive amounts of text, images, and PDFs.
The company views tools like the Chatgpt button as essential steps towards achieving artificial general intelligence, a concept that aims to match or exceed human intelligence in various tasks.
Last month, Openai unveiled an AI agent named Operator, claiming it can manage an online store based on photos of shopping lists, albeit only in the US preview version.
In a demonstration video released on Sunday, Openai showcased Deep Research analyzing the translation app market, stating that each task takes 5-30 minutes to complete with proper sourcing.
Openai highlighted that deep research targets experts in fields like finance, science, and engineering but can also be utilized for car and furniture purchases.
Leveraging Openai’s latest “reasoning” model, O3, deep research processes queries slower than traditional models and has a partially disclosed entity named O3-mini, a slimmed-down version of O3.
The full capabilities of the O3 model were outlined in the recent international AI safety report, prompting concerns from experts like Yoshua Bengio about the potential risks posed by AI advancements.
Deep surveys are accessible to Openai’s protia users in the US for $200 (£162 per month), with a monthly limit on queries due to processing constraints. Not available in the UK and Europe.
Andrew Rogoyski, director of an AI Research Institute affiliated with Sally University, cautioned about the potential dangers of blindly relying on deep search tools without conducting thorough verifications of their outputs.
“Knowledge-intensive AI faces a fundamental challenge. Human validation and verification are crucial to ensure the accuracy of machine analysis,” said Rogoyski.
There has been much talk about the potential for Poland’s economy to surpass that of the UK by 2030, but in some aspects, Poland is already ahead.
One such area is the digital ID card and driving license created by Poles, allowing them to access various public services through the mobile app mObywatel. Users must verify their identity through e-banking login, a digitally enabled physical ID card, or a special “trusted profile” online upon initial access.
With 8 million users, the mObywatel app enables Poles to create a digital ID, check demerit points on their driver’s license, review their car history, monitor local air quality, and find their polling place.
Rafał Sionkowski, a senior government official overseeing the app, emphasized the importance of keeping the core developer team within the public institution to ensure immediate public access to the digitized database.
As more EU countries develop similar apps in anticipation of the EU’s eIDAS 2.0 regulation on electronic identification, authentication, and trust services, significant progress is expected.
The regulation, set to be fully implemented by 2026 or 2027, establishes the legal framework for electronic identification systems that can be used across EU borders. Sionkowski noted that digital driving licenses can be presented in Germany and digital IDs in Spain for verification.
A digital version of your Polish driving license can be stored on your smartphone via an app. Photo: SOPA Images/LightRocket/Getty Images
Sionkowski mentioned plans to enhance the app with new features like notifying insurance companies of accidents and exploring its potential in verifying age online and assisting vulnerable groups in accessing public services.
He stressed the importance of focusing on services that people use, highlighting the value added through features like air quality monitors for local readings.
Privacy lawyer Wojciech Kulikki advocated for adhering to strict privacy principles while adding service features to the app. He cautioned against intrusive features like unauthorized location tracking.
Citizens could have more control over their data either through open-source app development for independent oversight or by checking data accessed by other government departments.
Janusz Ciezynski, a former digital minister, noted the smoother rollout of the app in Poland compared to the UK due to the presence of physical ID cards, quelling concerns about privacy infringements.
Ciezynski expressed enthusiasm for incorporating more public services into a single app, envisioning benefits for disaster-affected areas with quick access to funds through virtual payment cards.
Paleoanthropologists have characterized the properties of rough stone materials selected and used by early Pleistocene tool makers at Acheulean sites on the Ethiopian plateau between 16 and 1 million years ago.
Hand ax made from sidestrike flakes (ac) and kombewa flakes (df) from Melka Wakena, Ethiopia. Image credit: Tegenu Gossa & Erella Hovers, doi: 10.1007/s12520-024-02072-8.
The Hebrew University of Jerusalem said: “The two earliest stone tool assemblages, Oldowan and Acheulean, are now known to have emerged and proliferated within the East African Rift Valley.” Professor Herrera Hovers And my colleagues.
“The Oldowans (beginning 2.6 million years ago) introduced the obligatory use of percussion techniques, using various types of lithic supports (cobbles, nodules, pebbles, etc.) as percussion instruments to produce simple flakes. It represents a fundamental technological advance.
“The emergence of the Atyurs 1.75 million years ago probably marked major biological and behavioral changes in hominin lifestyles, often associated with improved cognitive abilities.”
In their study, the authors focused on: Melka Wakenaan early Achurian archaeological complex located in the south-central Ethiopian Plateau, at an altitude of 2,300 to 2,350 meters above sea level.
The site consists of several areas within a radius of approximately 2 km along the western bank of the Wabe River.
Preliminary investigations revealed the remains of animals, including 15 species of large vertebrates, some of which had human traces.
“Melka Wakena is one of the earliest known sites of high-altitude human habitation,” the researchers said.
“Evidence suggests that early humans made strategic choices based on factors such as rock compatibility, durability, and efficiency.”
“Melka Wakena's unique high-altitude setting provides valuable insight into how early humans adapted to their difficult environment.”
Scientists used advanced digital imaging techniques such as 3D scanning and photogrammetry to create highly detailed models showing the effects of stone tool use.
These models enable accurate analysis of wear patterns and surface changes and reveal that raw material properties have a significant influence on these changes, even under identical use conditions.
The findings suggest that early humans carefully evaluated the properties of materials when making tools, demonstrating a remarkable degree of technological foresight and adaptability.
“Our findings suggest that early humans were not simply picking up stones at random,” said Professor Hovers.
“They were making complex decisions about which materials were best for predictable needs, demonstrating a high degree of pre-planning and cognitive sophistication.”
E. Paisan others. 2025. Exploring early Acheulean technological decision-making: A controlled experimental approach to raw material selection for percussion crafts in Melka Wakena, Ethiopia. PLoS ONE 20 (1): e0314039;doi: 10.1371/journal.pone.0314039
Reports indicate that an artificial intelligence tool hosted by Amazon to enhance recruitment for the UK Ministry of Defense is potentially exposing defense workers to public identification risks. This information comes from a government evaluation.
The data utilized by automated systems in tailoring defense job advertisements to attract diverse candidates through inclusive language includes details like service member names, roles, and emails, and is stored by Amazon in the United States. A government document released for the first time today indicates that there is a risk of data breaches that could lead to the identification of defense personnel.
Although the risk has been classified as “low,” the Defense Department assured that there are “strong safeguards” in place by suppliers Textio, Amazon Web Services, and threat detection service Amazon GuardDuty.
The government acknowledges several risks associated with the use of AI tools in the public sector, as highlighted in a series of documents released to enhance transparency around algorithm use in central governments.
Ministers are advocating for the use of AI to enhance the UK’s economic productivity and deliver better public services. Safety measures are emphasized to mitigate risks and ensure resilience.
The UK government is collaborating with Google and Meta to pilot AI in public services. Microsoft is also offering its AI-powered Copilot system to civil servants, aligning with the government’s ambition to adopt a more startup-oriented mindset.
Some of the identified risks and benefits of current central government AI applications include:
Potential generation of inappropriate lesson material using a Lesson planning tool powered by AI, assisting teachers in customizing lesson plans efficiently.
Introduction of a chatbot to address queries concerning child welfare in family court, providing round-the-clock information and reducing wait times.
Utilization of a policy engine by the Ministry of Finance to model tax and benefit changes accurately.
Potential negative impact on human decision-making caused by excessive reliance on AI users in food hygiene inspections, leading to inconsistent scoring of establishments.
These disclosures will be documented in the expanded Algorithm Transparency Register, detailing information about 23 central government algorithms. Some algorithms with bias indications, like those in the Department for Work and Pensions welfare system, are yet to be recorded.
Imagine being able to visualize every aspect of our bodies, from our genes to the smallest cells that make up our organs. Scientists are now working on creating a comprehensive directory known as the Human Cell Atlas.
Think of it as a GPS for cells in the body, containing information on how cells evolve over time. This groundbreaking study, spanning 40 research papers, could potentially unravel major scientific mysteries such as bone formation, arthritis, and Crohn’s disease development.
Researchers have already gathered data on over 100 million cells from more than 10,000 individuals, with hopes of incorporating this information in the final atlas, which could potentially include billions of cells.
Professor Sarah Teichmann, the founding co-chair of the Human Cell Atlas and a researcher at the Cambridge Stem Cell Institute, stated, “This new level of insight into specific genes, mechanisms, and cell types within tissues lays the foundation for more accurate diagnosis, innovative drug discovery, and advanced regenerative medicine approaches.”
One significant milestone in the project focuses on the digestive system. By compiling data from 1.6 million cells, researchers have created an intricate map of intestinal cells, detailing their types, locations, and interactions within the body – the most detailed map of its kind.
Utilizing sophisticated techniques such as computer analysis and artificial intelligence (AI), researchers isolated individual cells and identified them, demonstrating the potential for AI in disease diagnosis.
A new approach similar to a “reverse image search” is being employed by researchers to aid in disease diagnosis, potentially revolutionizing the field.
Two striking images from the study stand out:
Image credit: Nathan Richoz University of Cambridge
This captivating image displays a magnified view of lung tissue, shedding light on the body’s response to COVID-19.
While it may resemble an artistic beach scene, this image actually depicts the intricate structures of the small intestine, a crucial organ in the digestive process.
A pebble with holes excavated from the ruins of Nahal Ein Geb II. Could be an ancient spindle whorl
Laurent Davin
A 12,000-year-old set of perforated pebbles unearthed in northern Israel may be the oldest known hand-spun whorls. This weaving technique may have ultimately helped inspire the invention of the wheel.
The whorl at the bottom of the spindle, which acted as a flywheel, allowed people to efficiently spin natural fibers into yarn and yarn to make clothing and other textiles. Newly discovered stone tools represent early axle-based turning technology, thousands of years before the first carts appeared, researchers say. Talia Yashuv at the Hebrew University of Jerusalem.
“If you look back at the discovery of the first car wheel 6,000 years ago, it didn't just come out of nowhere,” she says. “It's important to look at the functional evolution of how transportation and wheels have evolved.”
Yashuv and her colleagues leoa grossmanAlso at Hebrew University in Jerusalem, researchers studied 113 partially or fully perforated stones at the site of Nahal Ein Geb II, an ancient village just east of the Sea of Galilee. Archaeologists have been discovering these chalky, primarily limestone artifacts since 1972. It was probably made from raw pebbles found along the nearby shore.
The 3D scan revealed that the hole had been drilled part way from both sides using a flint hand drill, but unlike modern drills, it remained in the shape of a narrow, twisted cone, Yashuf said. says. A hole 3 to 4 centimeters in diameter usually passes through the center of gravity of the pebble.
Drilling holes from both sides would have balanced the stone and made it more stable to rotate, Yaszhu said. Some of the partially perforated stones had off-center holes, suggesting they may have been discarded by mistake.
Yashov said the team suspected that the stones, which weigh an average of 9 grams, would be too heavy and “ugly” to be beads, and too light and fragile to be used as fishing weights. Their size, shape, and balance around the hole led researchers to believe that these artifacts were spindle-like whorls.
To test their hypothesis, the researchers created a replica whorl using nearby pebbles and a flint drill. then they asked yonit crystalWhy not try your hand at flax spinning with a traditional craftsman?
“She was really surprised at how well they worked, because they weren't perfectly circular,” Yaszhu says. “But in reality, all you have to do is place the hole in the center of the mass, and it's balanced and works.”
If the stone is indeed a spiral, it could be the oldest known rotating spiral, she says. 1991 study of bone and antler artifacts She found what are likely 20,000-year-old whorls, but added that the researchers who examined them suggested the fragments were probably decorative accents on clothing. Still, people may have been using whorls even earlier, using wood and other biological materials that would have degraded.
The discovery suggests that people were experimenting with spinning techniques thousands of years before they invented potter's wheels and cart wheels about 5,500 years ago, and perhaps the whirlpool helped lead to those inventions. Yaszhu says it's possible.
carol cheval But a professor at the University of the Cote d'Azur in Nice, France, isn't so sure. She explains that the whorl works more like a spinning top than a wheel.
And while the artifact is very likely a whorl, the study lacks microscopic data that would reveal any traces of use, as the threads may have made their mark on the stone over time. , says Cheval.
Trace analysis is “beyond the scope” of the current study, Yashuv said.
Ideally, researchers studying ancient whorls would be skilled at spinning themselves, but the study authors said this was not the case. “It really changes the way you think about archaeological discoveries,” she says.
Argentine archaeologists analyzed 21,000-year-old fossil remains bearing cut marks belonging to a specimen of the extinct glyptodon. NeosclerocalyptusThe discovery, made in the northeast of the Pampean region, on the banks of the Reconquista River, adds new insights into the earliest human settlements in southern South America, and in particular into the interactions between humans and local megafauna in the Pampean region during the last glacial period.
Paleo-Indians hunting a Glyptodon, a relative of the armadillo that lived in the Pleistocene era, by Heinrich Harder, 1920.
The timing of early human occupation in South America is a topic of intense debate and is highly relevant to studies of early human dispersal across the Americas and the potential role of humans in the end-Pleistocene large mammal extinction.
This debate is hampered by a general lack of direct archaeological evidence for early human presence or human-animal interactions.
In the current study, Dr Mariano del Papa from the National University of La Plata and his colleagues found evidence of butchery in Pleistocene mammal fossils discovered on the banks of the Reconquista River in the northeastern Pampeo region of Argentina.
The fossils examined by the team were: Neosclerocalyptus Glyptodon, a giant relative of the armadillo.
Cut marks on the pelvis, tail and parts of the armour matched known marks made by stone tools.
The placement of these marks was consistent with a slaughter procedure that targeted densely packed areas of meat.
“Radiocarbon dating has dated these fossils to approximately 21,000 years ago, approximately 6,000 years older than any other known archaeological evidence from southern South America,” the researchers said.
Distribution of cut marks on the tail vertebrae NeosclerocalyptusImage courtesy of Del Papa others., doi: 10.1371/journal.pone.0304956.
The results are consistent with other recent studies showing early human presence in the Americas more than 20,000 years ago.
These fossils also represent some of the earliest evidence of human interaction with large mammals shortly before many of them went extinct.
“Our findings contradict the established dating framework for the earliest human occupation of southern South America, which was proposed to date back to 16,000 years ago,” the scientists said.
“Surprisingly, Recent Research “We now have reliable evidence of human habitation in Patagonia 17,300 years ago, indicating an even earlier initial settlement of southern South America.”
“Although traditional settlement models tend to support a later human migration into southern South America, we cannot exclude the possibility that humans were present and had associated cultural evidence much earlier than 16,000 years ago.”
“In this context, our findings support the growing body of archaeological evidence indicating an early human settlement in the Americas, especially the Southern Corn Islands.”
M. Delpapa others2024. Artificial cut marks on the bones of an extinct megafauna discovered in the Pampean region (Argentina) during the Last Glacial Maximum. PLoS One 19 (7): e0304956; doi: 10.1371/journal.pone.0304956
IBlocky World Chipotle Burrito BuilderIn Chipotle, players don the uniforms of the Tex-Mex restaurant chain and make burritos for virtual customers. Available toppings are taken from Chipotle’s real-world menu; shirts and caps feature the Chipotle logo. And when the game launched two years ago, the first 100,000 players earned “Burrito Bucks” to use in their burritos. Chipotle website.
after that Hyundai Mobility Adventure You can test drive models made by Korean manufacturers. Samsung Galaxy Station Here’s a mockup of the company’s latest smartphone designed to help travel to extraterrestrial worlds. Telefonica Town The challenge is to climb an assault course made from products featured in the telecommunications giant’s catalog. Vans World They just hand you a skateboard so you can do a few kickflips in a park plastered with shoe companies’ logos.
These are just a few of the corporate theme parks available. Robloxis one of the world’s most popular online video game platforms, with an average of 77 million players per day earlier this year, and is especially popular with children and younger players (58% of users self-reported as being under 16 years old). The end of last year), Roblox lets you explore fantastical virtual worlds, jumping over obstacles, finding hidden collectibles, and role-playing different tasks just like a kid would on the playground.
But the platform’s biggest selling point is its basic development tools, which allow anyone with little to no computer knowledge to create and share their own video games. Though this toolset is limited by design, it has attracted many people over the past few years, and not just aspiring game developers. This toolset has made Roblox a favorite playground for corporate advertisers, who use the development tools to create branded Roblox games to share with the game’s millions of players.
These advergames (advertisements presented in the form of video games) typically sprinkle corporate branding onto a set of game mechanics simple enough for Roblox’s younger player base. Broader suspicions Criticism that Roblox does not adequately protect children (which the company denies) has led to companies rushing to develop ad-supported games. Brands from Walmart to Wimbledon, McDonald’s to Gucci, Nike to the BBC have launched ad-supported games on the platform. Some have garnered hundreds of thousands of hits, others tens of millions. Seeking more brand involvement By promoting its large, young user base as a major attraction in a competitive advertising market.
An action shot from Vans World, where the company built a virtual skatepark in Roblox complete with footwear messaging. Photo: Vans / Roblox
“In the context of the attention economy, where consumers are exposed to hundreds, even thousands, of ads a day, capturing and maintaining attention is crucial,” says Yusuf Ochi, associate professor of marketing at Bayes Business School, City, University of London. “We are exposed to thousands of ads every day, many of which we don’t remember. Advagames circumvent these filters more effectively by integrating brand messaging into games.”
Öç’s own research has found that ads that utilize interactive features like touching, swiping, and tilting a phone screen can influence consumer preferences and purchase intent. Roblox allows brands to bring these interactive elements into a ready-made, engaging space.
“Roblox’s popularity with a younger demographic opens up new avenues for us to reach and engage the next generation of consumers in a sector where we’re already investing heavily,” said Robert Jan van Dormael, vice president of marketing for consumer audio at Samsung-owned Harman.
JBL, one of Harman’s hi-fi brands, released an official Roblox game in February, where players can collect audio snippets and arrange them into custom tracks, explore pastel-colored worlds and collect virtual currency to spend on cosmetic headphones and portable speakers, all accurately modeled after real-life JBL products. Since its release, it has attracted 1.4 million players, with average playtime over six minutes and engagement metrics orders of magnitude higher than other games. A few seconds A person typically spends an hour reading a social media post…
Bill Gates argues that artificial intelligence will assist, not hinder, in achieving climate goals, despite concerns about new data centers depleting green energy supplies.
The philanthropist and Microsoft co-founder stated that AI could enhance technology and power grids’ efficiency, enabling countries to reduce energy consumption even with the need for more data centers.
Gates reassured that AI’s impact on the climate is manageable, contrary to fears that AI advancements might lead to increased energy demand and reliance on fossil fuels.
“Let’s not exaggerate this,” Gates emphasized. “Data centers contribute an additional 6% in energy demand at most. But it’s likely around 2% to 2.5%. The key is whether AI can accelerate the reduction to 6% or beyond. The answer is, ‘Definitely.’
Goldman Sachs estimates that AI chatbot tool ChatGPT’s electricity consumption for processing queries is nearly ten times more than a Google search, potentially causing carbon dioxide emissions from data centers to double between 2022 and 2030.
Experts project that developed countries, which have seen energy consumption decline due to efficiency, could experience up to a 10% rise in electricity demand from the growth of AI data centers.
In a conference hosted by his venture fund Breakthrough Energy, Gates told reporters in London that the additional energy demand from AI data centers is likely to be offset by investments in green electricity, as tech companies are willing to pay more for clean energy sources.
Breakthrough Energy has supported over 100 companies involved in the energy transition. Gates is heavily investing in AI through the Gates Foundation Trust, which has allocated about a third of its $77 billion assets into Microsoft.
However, Gates’ optimism about AI’s potential to reduce carbon emissions aligns with peer-reviewed papers, suggesting that generative AI could significantly lower CO2 emissions by simplifying tasks like writing and creating illustrations.
AI is already influencing emissions directly, as demonstrated by Google using deep learning techniques to reduce data center cooling costs by 40% and decrease overall electricity usage by 15% for non-IT tasks.
Despite these advancements, concerns remain about the carbon impact of AI, with Microsoft acknowledging that its indirect emissions are increasing due to building new data centers around the world.
Gates cautioned that the world could miss its 2050 climate goals by up to 15 years if the transition to green energy is delayed, hindering efforts to decarbonize polluting sectors and achieve net-zero emissions by the target year.
He expressed concerns that the required amount of green electricity may not be delivered in time for the transition, making it challenging to meet the zero emissions goal by 2050.
Gates’ warning follows a global report indicating a rise in renewable energy alongside fossil fuel consumption, suggesting that meeting climate goals requires accelerated green energy adoption.
As a human, you will play a crucial role in identifying whether a photo or video was created using artificial intelligence.
Various detection tools are available for assistance, either commercially or developed in research labs. By utilizing these deepfake detectors, you can upload or link to suspected fake media, and the detector will indicate the likelihood that it was generated by AI.
However, relying on your senses and key clues can also offer valuable insights when analyzing media to determine the authenticity of a deepfake.
Although the regulation of deepfakes, especially in elections, has been slow to catch up with AI advancements, efforts must be made to verify the authenticity of images, audio, and videos.
One such tool is the Deepfake Meter developed by Siwei Lyu at the University at Buffalo. This free and open-source tool combines algorithms from various labs to help users determine if media was generated by AI.
The DeepFake-o-meter demonstrates both the advantages and limitations of AI detection tools by rating the likelihood of a video, photo, or audio recording being AI-generated on a scale from 0% to 100%.
AI detection algorithms can exhibit biases based on their training, and while some tools like DeepFake-o-meter are transparent about their variability, commercial tools may have unclear limitations.
Lyu aims to empower users to verify the authenticity of media by continually improving detection algorithms and encouraging collaboration between humans and AI in identifying deepfakes.
audio
A notable instance of a deepfake in US elections was a robocall in New Hampshire using an AI-generated voice of President Joe Biden.
When subjected to various detection algorithms, the robocall clips showed varying probabilities of being AI-generated based on cues like the tone of the voice and presence of background noise.
Detecting audio deepfakes relies on anomalies like a lack of emotion or unnatural background noise.
photograph
Photos can reveal inconsistencies with reality and human features that indicate potential deepfakes, like irregularities in body parts and unnatural glossiness.
Analyzing AI-generated images can uncover visual clues such as misaligned features and exaggerated textures.
An AI-generated image purportedly showing Trump and black voters. Photo: @Trump_History45
Discerning the authenticity of AI-generated photos involves examining details like facial features and textures.
video
Video deepfakes can be particularly challenging due to the complexity of manipulating moving images, but visual cues like pixelated artifacts and irregularities in movements can indicate AI manipulation.
Detecting deepfake videos involves looking for inconsistencies in facial features, mouth movements, and overall visual quality.
The authenticity of videos can be determined by analyzing movement patterns, facial expressions, and other visual distortions that may indicate deepfake manipulation.
CDC Director Mandy Cohen emphasized the importance of utilizing tools and guidelines to help individuals identify places to stay cool when air conditioning is not available, recognize symptoms of heat illness, and properly manage medications. Cohen highlighted the significance of understanding how drugs interact with heat during a press conference on Monday.
“While heat can impact our health, it is crucial to remember that heat-related illness and death are preventable,” Cohen stated.
Heat-related deaths outnumber those caused by other extreme weather events such as floods, hurricanes, and tornadoes in the United States each year. The record-breaking heat experienced last summer highlighted the threat of scorching temperatures, particularly in the South and Southwest regions of the country.
NOAA officials expressed optimism that the new resources will assist communities in preparing for the upcoming summer season. The agency anticipates above-average temperatures in May and June across the United States, indicating another hot summer ahead.
“It is never too early to start preparing for heat-related challenges,” emphasized NOAA Administrator Rick Spinrad during a briefing.
NOAA’s HeatRisk tool categorizes heat risks on a scale from 0 (green) to 4 (magenta), with 4 indicating extreme and/or prolonged heat impacts. The tool considers factors such as maximum and minimum temperatures as well as the combined effects of heat during both day and night. It is tailored to provide location-specific heat outlooks as environmental conditions vary from one place to another.
The forecast also includes historical data to provide context on the predicted temperatures relative to past records during the same time of year.
NOAA National Weather Service Director Ken Graham highlighted that the HeatRisk tool can assist individuals in making informed decisions about outdoor activities based on the heat risk level. The tool aims to complement heat watches and warnings issued by government agencies by offering additional context for users.
The initial prototype of the HeatRisk tool was developed for California by the National Weather Service in 2013 and expanded to include Western states in 2017. It is currently available as a trial tool across the continental United States.
Members of the public are encouraged to submit feedback on the tool by September 30th to the National Weather Service.
Pessimism is a dirty word in climate policy circles. There are good reasons for this. Especially because while optimism can encourage positive change, assuming the worst can paralyze us and prevent us from taking action. But when it comes to climate modeling, a certain amount of negativity can be a good thing.
The Intergovernmental Panel on Climate Change is already working on various models and pathways to assess how to limit warming to 1.5°C and how to ensure that carbon emissions continue unabated or experience many possibilities in between. I use it to hedge my bets. These pathways are backed by thousands of scientific papers, tons of data, and the brains of the world’s climate scientists, but like all models, they are built on assumptions.
One of the key assumptions in the scenario of keeping temperature rise below 1.5°C is that the technology to remove carbon dioxide from the atmosphere will be rapidly perfected in the near future. This is not an unreasonable prediction, given human ingenuity and strong incentives to do so. But incorporating carbon capture technology into these models is like declaring that winning the lottery will balance the household budget. If you can’t reduce your spending to an affordable level, you better hope that a big prize is on the way.
As the two articles in this issue demonstrate, this is a dangerous approach. A detailed analysis of geological carbon storage plans shows that it is at least very unlikely, if not impossible, to meet the levels envisioned for many 1.5°C pathways. (“Our plans to tackle climate change with carbon storage add up”). The chances of winning the lottery don’t seem that high. On the other hand, we also received an unexpected carbon bill in the form of melting Arctic permafrost, releasing more greenhouse gases than previously accounted for. Frozen soil is now a major net source of greenhouse gases (see “Frozen soil is now a major net source of greenhouse gases”).
While these revisions in our understanding of climate change are entirely expected and to be welcomed, they do signal that the challenges we face over the next decade will only get more difficult. . Rather than narrowing down climate models until the numbers roughly match the 1.5°C goal, perhaps it would be better to take a more pessimistic outlook and accelerate efforts to limit the damage.
Do you notice your muscles becoming more rigid and harder to manage as you age? A new ‘Atlas of Aging’ has been developed to explain why this happens and to provide potential treatments to prevent it. Additionally, it may lead to legal action.
Focusing on the effects of natural aging, this atlas delves into the intricate changes that occur in muscle tissue at the cellular and molecular levels as we grow older. It also highlights how our muscles actively combat the aging process, potentially aiding in the development of new treatments to enhance the aging body.
As we age, our muscles can weaken, making everyday activities like standing and walking more challenging. However, the underlying causes of this decline are not fully understood. Frailty can lead to an increased risk of falls, reduced mobility, and loss of independence.
Lead author, Dr. Sarah Teichman from the Wellcome Sanger Institute, states that these insights into healthy skeletal muscle aging are empowering researchers worldwide to explore various strategies to combat inflammation, promote muscle regeneration, maintain neural connections, and more.
Longevity expert Andrew Steele emphasizes the importance of understanding the cellular changes that contribute to the loss of physical strength as we age. He underscores the potential of this research to develop therapeutic interventions that support healthier aging in future generations.
The creation of the atlas of aging muscle involved utilizing advanced imaging and single-cell sequencing techniques to analyze skeletal muscle samples from 17 adult donors aged between 20 and 75. The findings shed light on gene activity related to protein production and revealed how muscle fibers age at different rates.
Age-related loss of primary fast-twitch muscle fibers is mitigated by the body’s ability to enhance the properties of remaining fibers and rebuild connections between weakened nerves and aging muscles. This understanding can potentially inform strategies to maintain strength and independence as we grow older.
To learn more about the experts involved in this research, Dr. Andrew Steele, a scientist, author, and presenter, has authored “Ageless: The new science of growing older without getting older.” Combining his background in physics with biology, Steele’s work focuses on deciphering human DNA at the Francis Crick Institute in London.
Read more:
What happens to my body as I get older?
9 simple science-backed changes to reverse your biological age
Groundbreaking discovery of anti-aging cells could help people stay young for longer
OpenAI on Thursday announced a tool that can generate videos from text prompts.
The new model, called Sora after the Japanese word for “sky,” can create up to a minute of realistic footage that follows the user’s instructions for both subject matter and style. The model can also create videos based on still images or enhance existing footage with new material, according to a company blog post.
“We teach AI to understand and simulate the physical world in motion, with the goal of training models that help people solve problems that require real-world interaction.” says the blog post.
One video included among the company’s first few examples was based on the following prompt: Movie trailer featuring the adventures of a 30-year-old astronaut wearing his red woolen knitted bike in his helmet, blue sky, salt desert, cinematic style shot on 35mm film, vibrant colors .”
The company announced that it has opened up access to Sora to several researchers and video creators. According to the company’s blog post, experts have “red-teamed” the product and implemented OpenAI’s terms of service, which prohibit “extreme violence, sexual content, hateful images, likenesses of celebrities, or the IP of others.” We will test whether there is a possibility of evasion. The company only allows limited access to researchers, visual artists and filmmakers, but CEO Sam Altman took to Twitter after the announcement to answer questions from users about a video he said was created by Sola. posted. The video contains a watermark indicating that it was created by AI.
The company debuted its still image generator Dall-E in 2021 and its generated AI chatbot ChatGPT in November 2022, quickly gaining 100 million users. His other AI companies have also debuted video generation tools, but those models could only generate a few seconds of footage that had little to do with the prompt. Google and Meta said they are developing a video generation tool, although it is not publicly available. on wednesday, announced the experiment We’ve added deeper memory to ChatGPT to remember more of your users’ chats.
OpenAI told the New York Times how much footage was used to train Sora, except that the corpus includes videos that are publicly available and licensed from copyright holders. He also did not reveal the source of the training video. The company has been sued multiple times for alleged copyright infringement in training generative AI tools that digest vast amounts of material collected from the internet and mimic the images and text contained in those datasets. .
New research has found that diabetes may be diagnosed with just a short audio recording from a mobile phone.
Scientists can determine whether someone has diabetes with nearly 90% accuracy using just a 6-10 second audio sample and basic health data such as age, gender, height, and weight I created an AI model.
Klick Labs recruited 267 people for the study, including some who had already been diagnosed with type 2 diabetes.
Each subject was asked to record a phrase on their phone six times a day for two weeks, and the team used AI to analyze more than 18,000 samples to determine the acoustic differences between diabetics and non-diabetics. I looked into it.
These included changes in pitch caused by type 2 diabetes that are imperceptible to the human ear.
This model had an accuracy rate of 89% for women and 86% for men.
Study author Jaycee Kaufman said the results could “change” the way we screen for diabetes.
More than 90% of adults with diabetes in the UK have type 2 diabetes, but many go undetected for years as symptoms may be systemic or absent. I am.
Testing for this disease usually requires a visit to a general practitioner and urine and blood tests.
“Current detection methods can be time-consuming, travel-intensive, and costly,” Kaufman said.
“Voice technology has the potential to completely remove these barriers.”
Read more science and technology news: The rise and fall of BlackBerry Debunking myths about the influenza vaccine NASA reveals ‘unbelievable’ asteroid discovery
Previous research has shown that audio recordings can be used in conjunction with AI to diagnose other diseases. Including new coronavirus infection.
Klick Labs believes this technology can also diagnose conditions such as prediabetes and hypertension.
This peer-reviewed study was published in the Mayo Clinic Proceedings journal.
This website uses cookies so that we can provide you with the best user experience possible. Cookie information is stored in your browser and performs functions such as recognising you when you return to our website and helping our team to understand which sections of the website you find most interesting and useful.
Strictly Necessary Cookies
Strictly Necessary Cookie should be enabled at all times so that we can save your preferences for cookie settings.
