Tag Archives: current

Brian Johnson’s Current Projects: What He’s Up to Now

Posted on by
Reply

New Scientist: Discover the latest in science, technology, health, and environmental news from expert journalists.

Feedback is New Scientist A go-to source for individuals keen on the latest developments in science and technology. Share your insights with us via email at feedback@newscientist.com.

Headline Highlights

Crafting great headlines can be challenging. The initial words may seem unusual, only to get even more unexpected as you read on. By the end, you question if you’re consuming reliable news or an experimental novel akin to James Joyce’s works.

For instance, a striking headline appeared in an online music magazine on November 29th in Stereogum:
“A shroom trip with immortal influencers and special guest Mr. Beast, DJed by Grimes.”
Don’t worry if it sounds confusing; I’ll clarify the context shortly.

Starting with Grimes: an artist known for her sci-fi-inspired music, her climate-focused 2020 album is titled Miss Anthropocene, while her debut, Gady Prime, pays homage (albeit with a misspelling) to Frank Herbert’s Dune.

Conversely, Brian Johnson is a tech mogul determined to achieve longevity, engaging in a variety of life-extending efforts. His regimen includes exercise, dietary modifications, and even experimenting with rapamycin, a drug typically prescribed to transplant patients (which he has ceased taking), alongside his bold ambition to upload his consciousness into an AI.

The narrative unfolds as Johnson consumes hallucinogenic mushrooms during a livestream, while undergoing numerous health assessments. Grimes was invited to provide musical entertainment during this session. Although YouTuber Mr. Beast did not participate, other noteworthy figures like Marc Benioff, CEO of Salesforce, and technology journalist Ashlee Vance were involved. If we were to explore the effects of psilocybin, ideally, a trusted companion and qualified therapist would be present. But Johnson clearly had a plan.

Details about the event can be found in a video online, which spans a little over five and a half hours. While feedback should have closely monitored this session, we recognize that unlike Johnson, we acknowledge our mortality and don’t want to spend our time recklessly.

Unexpected Inquiry

This may lead to a recurring segment in your feedback: “Questions You Never Considered.” Our reader Keith Edkins stumbled upon such an inquiry. To those attempting to investigate it, good luck!

Keith referenced a 2014 paper in Folia Parasitologica, known for its focus on parasites. Among them, Toxoplasma gondii, a single-celled organism that infects cats and is prevalent in many humans, might be linked to psychiatric disorders such as intermittent explosive disorder. The paper poses an intriguing question: “Is there a correlation between the prevalence of subclinical toxoplasmosis and the frequency of rhesus-negative patients with national traffic accident rates?”

As Keith noted, “With proper statistical handling, the answer seems to be ‘no.’” But what a question! Can anyone challenge it?

Graphic Anomalies

Descriptive graphics can often perplex. Throughout our extensive journey in science journalism, we’ve often sought clarity in complex figures shared by researchers. From convoluted flowcharts to intricately shaded bar graphs, confusion has been a constant.

However, a recent publication in Scientific Reports caught attention: it depicts a woman receiving a tin of biscuits. As our reader Jim Santo aptly remarked, “This is wild,” and it’s true. Released on November 19th, the study aims to establish an AI-driven system to assist in diagnosing autism spectrum disorders. While feedback doesn’t express particular opinions about the study itself, it has gathered significant interest.

Quickly downloading a copy after spotting discussions about the paper on social media, the primary concern raised was Figure 1, touted as “the overall behavior of the framework depicted as an infographic.” It’s astonishing enough to demand attention.

Centered in this graphic is a woman cradling a small child; intriguingly, her feet seem to be encased in concrete. The child points towards a speech bubble saying “Missing Values and Executive Functions.” To the right, another speech bubble displays “Historic Medical Frimbral and Environmental Features.”

Additional elements include pink blobs resembling deformed kidney beans, symbolizing “7 TOL Llne storee.” Also featuring is a so-called “Factor Fexsectern” and a mysterious spiked bicycle.

As noted in the magazine’s retraction, all of this was produced by AI. As we continued to observe the graphic, growing intrigue stirred within. There’s a mention of “Totalbottl” at its base, but the actual meaning remains elusive. As for the bicycle, perhaps someone was riding it.

Curiously, this Scientific Reports article could claim the title of one of the fastest withdrawals we’ve encountered. Typically, journals may take years to retract flawed studies. On December 3rd, Retraction Watch reported that numerous works attributed to psychologist Hans Eysenck might require retraction due to “questionable data” and other anomalies, including an outlandish assertion suggesting some individuals possess a “cancer-prone personality.” Interestingly, Eysenck passed away in 1997, illustrating the slow pace of rectifying such issues.

Have a story for feedback?

Share your insights by emailing Feedback at feedback@newscientist.com. Don’t forget to include your home address. You can find this week’s and past feedback on our website.

Source: www.newscientist.com

COP30: UN Climate Summit No Longer Meets Current Needs

Posted on by
Reply

Pablo Porciuncula/AFP via Getty Images

Consider having a life-threatening illness. All scientific assessments point to a definitive diagnosis and a grim prognosis. Yet, upon visiting a doctor, they fail to acknowledge the condition directly. After some brief small talk, they shake your hand and suggest scheduling your next appointment in a year.

No one would accept such a medical standard, yet this mirrors our approach to climate change. The 30th Conference of the Parties (COP) Summit wrapped up last weekend in Belém, Brazil. While notable strides have been made in addressing climate change, particularly with the 2015 Paris Agreement aimed at restricting temperature increases to below 1.5°C, this goal is largely unachieved. Nonetheless, it steers us towards reduced warming compared to what might have been without it.

However, it’s evident that the COP process is becoming inadequate for the challenges we face. As highlighted in our report on page 6, COP30 concluded without even mentioning fossil fuels—the primary driver of climate change—in the final document. Despite over 80 nations advocating for a roadmap toward a “post-fossil fuel transition,” this initiative faced resistance from oil-rich nations like Saudi Arabia, a former organizer of COP. The necessity of consensus within COP leaves us only with the promise of future discussions at COP31 in Turkey next year.

“
Nations advocating for climate action should prioritize solar power and battery technology.
“

This situation cannot persist, but changing the COP process will be a challenge. If we can’t advocate for an end to the fossil fuel era through scientific and political means, we must turn to technological and economic solutions.

Nations committed to climate action should concentrate on solar energy and battery technologies, providing the world with cheaper alternatives to oil and gas. Countries striving for a sustainable future might need to implement economic sanctions against those showing disinterest in progress. Whatever the course of action, simply saying “see you next year” is no longer a viable option.

Source: www.newscientist.com

Overall Positive: Tim Berners-Lee Talks About the Web’s Current State

Posted on by
Reply

Tim Berners-Lee in a rack at the CERN computer center

Maximilian Bryce/CERN

Tim Berners-Lee holds a comprehensive map of the internet on a single page, featuring around 100 blocks linked by various arrows. These blocks encompass blogs, podcasts, group messages, and abstract themes like creativity, collaboration, and clickbait, providing a unique depiction of the digital realm from the innovator of the World Wide Web.

“Most of them are good,” he remarked during our conversation at New Scientist‘s London office, reflecting on the web’s successes and failures. This map serves as a guide for others and a reminder that only a small fraction of the Internet is deemed detrimental to society. The top-left quadrant illustrates Berners-Lee’s concerns, with six blocks marked “Harmful,” including names like Facebook, Instagram, Snapchat, TikTok, X, and YouTube.

In the last 35 years, Berners-Lee’s creation has evolved from just one user (himself) to approximately 5.5 billion users, constituting about 70% of the global population. It has transformed communication and shopping, making modern life unimaginable without it. However, the list of emerging challenges continues to expand.

Issues like misinformation, polarization, and election interference have become staples of online discourse, contrasting sharply with Berners-Lee’s vision of a collaborative utopia. In his memoir, This is for Everyone, he reflects, “In the early days of the web, joy and wonder were abundant, but today’s online experience can induce just as much anxiety.”

It’s natural for the web’s architect to feel a sense of disappointment regarding humanity’s use of his creation, yet he remains hopeful for the future of the internet. As one of the foremost technology visionaries (with a plethora of accolades and honors), he shares insights on what went awry and how he envisions solutions.

Invention of the Web

The World Wide Web’s origin story hinges on being at the right place and time. In the late 1980s, Berners-Lee was part of the computing and networking sector at a U.S. company. At CERN, the particle physics lab near Geneva, Switzerland, he pondered over better document management methods.

Most systems forced users into rigid organizational structures and strict hierarchies. Berners-Lee envisioned a more flexible approach, permitting users to link documents freely. Hyperlinks existed for internal references, and the Internet was already available for file sharing—why not merge the two concepts? This simple yet transformative idea birthed the World Wide Web.

Although Berners-Lee had harbored the idea since 1989, he ultimately convinced his supportive supervisors to let him pursue it fully. Within months, he created a surge of developments that led to HTML—a programming language for web pages, HTTP—the protocol for transferring them, and the URL, the means to locate them. The final code consisted of just 9,555 lines, marking the web’s emergence by year’s end.The web was born.

“CERN was an extraordinary place to innovate the web,” he states. “Individuals from around the world, driven by a genuine need to communicate and document their experiences, came together there.”

The inaugural website was hosted on Berners-Lee’s work computer, adorned with a “Do Not Turn Off” sign and instructions for engaging with the web. More web servers emerged, leading to exponential growth: “In the first year, it grew tenfold; in the second year, another tenfold; and by the third, yet another tenfold.” He recalls, “Even then, I sensed we were onto something significant.”

Initially, most web pages were crafted by academics and developers, but soon, everyone began using them to share a wide array of content. Within a decade, the landscape blossomed into millions of websites, hundreds of millions of users, and the inevitable rise of dot-com ventures.

The Spice Girls with their website in 1997.

David Corio/Redferns

Despite the web’s immense potential for profit, Berners-Lee believed it should remain free and open to realize its full capabilities. This was a challenge, as CERN had legitimate grounds to claim royalties on the software being developed. Berners-Lee advocated for his superiors to release this technology openly, and by 1993, after much negotiation, the comprehensive source code of the Web was made available, complete with a disclaimer: CERN relinquishes all intellectual property rights to this code—the web will be royalty-free forever.

Early Days

For its initial years, the web flourished. Although there was a notorious stock market crash at the turn of the millennium, largely driven by speculative venture capital rather than the web itself, piracy was rampant, and malware was ever-present, the web was fundamentally open, free, and enjoyable. “People loved the web; they were simply happy,” Berners-Lee recounted in his memoir.

He captured the essence of this era, believing the web held the potential to foster new forms of collaboration among people. He coined the term “intercreativity” to describe the creative synergy of groups rather than individuals. Wikipedia, with around 65 million English pages edited by 15 million contributors, exemplifies what he envisioned for the web. He notably positions it on his map and describes it as “probably the best single example” of his aspirations.

However, the optimistic phase of the web was not to extend indefinitely. For Berners-Lee, the turning point came in 2016, marked by the Brexit vote and the election of Donald Trump. “At that moment, discussions arose about how social media could be manipulated to influence voters against their interests. In essence, the web became an instrument of manipulation driven by larger entities,” he shared.

Traditionally, political movements communicated their messages to the public openly, allowing for critique and discussion. However, by the mid-2010s, social media enabled “narrowcasting,” as Berners-Lee describes it, allowing political messages to be tailored into numerous versions for various audiences. This complicates tracking who communicated what and makes it harder to counter misinformation.

The extent of this microtargeting’s impact on elections remains debated. Numerous studies have tried to quantify how such messaging alters public opinion and voting behavior, generally uncovering only modest effects. Regardless, these trends contribute to Berners-Lee’s broader concerns about social media.

He emphasized that social media platforms are incentivized to maintain user engagement, which leads to the creation of “addictive” algorithms. “People are naturally drawn to things that evoke anger,” he states. “When social media feeds users misinformation, it’s more likely to garner clicks and ensnare users longer.”

Quoting author Yuval Noah Harari, he stipulated that creators of “harmful” algorithms should likewise be held accountable for their recommendations. “It’s particularly essential to undermine systems designed to be addictive,” Berners-Lee argues. He admits that imposing restrictions contradicts his usual free and open philosophy, viewing it as a last resort. Social media can unify individuals and disseminate ideas, yet it also poses unique risks that warrant change, as he specifies in his latest book. “This must evolve somehow.”

Nonetheless, he harbors an optimistic view of the web’s potential trajectory. While social media, despite its captivating nature, represents merely a fragment of the internet landscape, Berners-Lee contends that addressing these issues should be part of a broader strategy aimed at enhancing the web overall, with a focus on reclaiming digital sovereignty.

A Plan for Universal Web Access

To further this goal, Berners-Lee has dedicated the last decade to developing a new framework reinstating control with the individual. Presently, disparate internet platforms manage personal data. For instance, it’s challenging to share a video from Snapchat on Facebook or a post from LinkedIn to Instagram—the user can create this content, yet each company retains ownership.

Berners-Lee’s concept advocates for consolidating data into a singular data repository known as a pod (short for “personal online data store”), which the user controls, rather than having information dispersed across various platforms. This pod can hold everything from family images to medical records, with users determining what to share. This isn’t merely theoretical; he co-founded a company, Inrupt, that aims to bring this vision to life.

Berners-Lee using an early version of website and web browser invented at CERN in 1994

CERN

He is particularly enthusiastic about merging data wallets with artificial intelligence. For example, when searching for running shoes, current AI chatbots require detailed guidance to offer suitable recommendations. However, if an AI accesses a user’s data wallet, it can understand all past measurements, training history, and potentially spending behavior, leading to more accurate suggestions.

Berners-Lee advocates that AI should serve users, not large tech corporations. His goal isn’t to create individual AIs but to establish safeguards within software. Data wallets are part of the solution, along with an idea that AI should adhere to a kind of digital Hippocratic oath to avoid causing harm. He envisions AI acting as “your personal assistant,” providing tailored support.

While recommending appropriate running shoes may not address the web’s most pressing challenges, Berners-Lee possesses an exceptional ability to envision potential before others. Data wallets might seem mundane today, yet just decades ago, hyperlink-based document management systems were equally obscure. His passion for bettering the world drives him, as he believes enhancing the data ecosystem is crucial to achieving that goal.

All these developments suggest Berners-Lee envisions a fundamental shift for the web. He believes we must transition from an “attention economy,” dominated by competing clicks, to an “intention economy,” where users express their needs and companies—and AI—strive to fulfill them. “This is more empowering for the individual,” he asserts.

Such a transformation could redistribute power from tech giants to users. Some might think such a reversal unlikely, especially with the ongoing trends of tech dominance and the pervasive “doomscrolling” culture. However, Berners-Lee has a proven history of spotting opportunities others miss, and ultimately, he is the architect of the roadmap.

Topics:

Source: www.newscientist.com

Crested Diapsid Reptiles from the Central Triassic Challenge Current Theories of Wing Evolution

Posted on by
Reply

Paleontologists have identified a novel genus and species of Triassic derepanosauromorph diapsid showcasing remarkable appendages (not feathers or skin). This discovery is based on two exceptionally well-preserved skeletal structures and related specimens. Their research reveals that wings and hair-like extensions are not exclusive to birds and mammals.

Mirasaura Grabogeli In natural forest environments, insects are hunted. Image credit Gabriel Uguet.

Feathers and hair are intricate outer body appendages of vertebrates, serving essential functions such as insulation, sensory support, display, and facilitating flight.

The development of feathers and hair traces back to the ancestral lines of birds and mammals, respectively.

However, the genetic frameworks responsible for these appendages may have origins deeper within the amniotic lineage, encompassing various animal branches, including those of birds and mammals.

The Triassic reptile species outlined by Dr. Stephan Spiekman from the Staatliches Museum für Naturkunde Stuttgart and his collaborators featured unique appendages that could reach up to 15.3 cm (6 inches) in length along their backs.

Named Mirasaura Grabogeli, this peculiar creature inhabited Europe approximately 247 million years ago.

The species exhibited a superficially bird-like skull but was classified within the Diapsid group known as Drepanosauromorpha.

Anatomy of Mirasaura Grabogeli. Image credit: Spiekman et al., doi: 10.1038/s41586-025-09167-9.

Discovered in northeastern France in the 1930s, Mirasaura Grabogeli comprises 80 specimens featuring two well-preserved skeletal structures with isolated appendages and preserved soft tissues. Recent preparations have led to its identification.

“This enabled the connection between the summit and skeleton,” the paleontologist noted.

“The tissue preserved within the appendages contains melanosomes (pigment-producing cells located in skin, fur, and feathers), resembling those found in feathers more closely than in reptilian skin or mammalian hair, yet lacking the typical branching pattern of feathers.”

“These observations suggest that such complex appendages might have evolved among reptiles prior to the emergence of birds and their closest relatives, potentially offering new insights into the development of feathers and hair.”

Given the characteristics of the appendages observed in Mirasaura Grabogeli, we dismissed their roles in flight or camouflage, proposing instead a possible role in visual communication (signaling or predator deterrence).

The team’s research paper was published today in the journal Nature.

____

SNF Spiekman et al. Triassic diapsids reveal early diversification of skin appendages in reptiles. Nature Published online on July 23, 2025. doi:10.1038/s41586-025-09167-9

Source: www.sci.news

The neighborhood’s current thoughts on your dementia risk

Posted on by
Reply

Living in different areas can greatly impact your health. Various factors, such as the environment, income, and overall living conditions, can play a role in affecting your long-term well-being. Recent studies suggest that these factors may also influence your chances of developing dementia.

A new study published in the Journal Neurology revealed that individuals residing in disadvantaged neighborhoods are more than twice as likely to develop dementia compared to those in wealthier areas.

Conducted by Professor Pankaja Desai at Rush University in Chicago, the study involved over 6,800 participants aged 65 and older from four nearby communities. The research found that individuals in the most disadvantaged areas had a 22% risk of developing dementia, whereas those in more privileged areas had only an 11% risk.

Even after adjusting for factors like age, gender, and education, the study observed that individuals in disadvantaged neighborhoods were more than twice as likely to develop Alzheimer’s disease. This connection was determined using the Social Vulnerability Index (SVI), which incorporates various socio-economic factors to assess neighborhood-level risk.

Furthermore, individuals in disadvantaged areas experienced a faster decline in cognitive function as they aged, regardless of an Alzheimer’s disease diagnosis. This emphasizes the impact of community-level factors on dementia risk.

According to Desai, addressing neighborhood-level social characteristics is crucial in reducing the risk of Alzheimer’s and planning efficient health services. The study also highlighted disparities in dementia risk among different racial groups, indicating the importance of considering community factors in dementia care.

While the study’s focus was on Chicago neighborhoods and may not be universally applicable, the findings underscore the link between neighborhood disadvantage and dementia risk. Ultimately, the study emphasizes the significance of environmental factors in brain health.

About our experts

Dr. Pankaja Desai is an assistant professor at the Rush Institute for Healthy Aging and serves as the management director of Rush Bioinformatics and Biostatistics Core. Her research has been featured in publications such as American Journal of Health Behavior and Alzheimer’s Disease and Dementia.

read more:

Source: www.sciencefocus.com

AI researchers doubt that current models will result in AGI

Posted on by
Reply

Many AI companies say their models are on the path to artificial general information, but not everyone agrees

Manaure Quintero/AFP via Getty Images

Tech companies have argued that simply expanding their current AI models will lead to artificial general information (AGI). However, the performance of modern models is high, so AI researchers doubt that today's technology will lead to tighter systems.

In a survey of 475 AI researchers, approximately 76% of respondents said they were “impossible” or “very unlikely” to succeed in achieving AGI by expanding their current approach. The survey results are part of a Report by the Society for Progress in Artificial Intelligence, an International Association for Science based in Washington, DC.

This is a noticeable shift in the “need to scale” attitude that has spurred high-tech companies since the launch of the generative AI boom in 2022. Since then, most of the cutting-edge achievements have been trained by increasing the amount of data, which has resulted in improved performance. However, they appear to be stagnant with their latest releases, showing only progressive changes in quality.

“The enormous investment in scaling seemed to be constantly left behind, accompanied by comparable efforts to understand what was going on.” Stuart Russell He was a member of the panel that compiled the report at the University of California, Berkeley. “I think it began to be clear to everyone that about a year ago the benefits of scaling in the traditional sense took away the layers.”

Nevertheless, tech companies plan to spend collectively Estimated $1 trillion Support AI ambitions with data centers and chips for the next few years.

Hype about AI technology may explain why 80% of survey respondents said their current perceptions of AI capabilities were not consistent with reality. “Systems that are declared to match human performance, such as coding problems and mathematical problems, are making painstaking mistakes.” Thomas Neetteric He contributed to the report at Oregon State University. “These systems are extremely useful tools to support research and coding, but they do not intend to replace human workers.”

AI companies have recently focused on what is called inference time scaling, which takes longer for AI models to use more computing power and process queries before responding. Arvind Narayanan At Princeton University. However, he says that this approach is “a unlikely to become a silver bullet” to reach the AGI.

High-tech companies often describe AGI as their ultimate goal, but the very definition of AGI is unstable. There is Google DeepMind explained It is a system that can outperform all humans in a series of cognitive tests, and Huawei has Proposed To reach this milestone, we need a body that allows AI to interact with its environment. Internal reports for Microsoft and Openai It is listed Considering that AGI can only be achieved if Openai develops a model that can generate $100 billion in profits.

topic:

  • artificial intelligence/
  • Computing

Source: www.newscientist.com

First-ever imaging of a novel form of magnetism: alternating current magnetism

Posted on by
Reply

Alternating current magnetism is a unique form of magnetic ordering in which small magnetic components align antiparallel to their neighbors, but the structure housing each element is rotated relative to its neighbors. . Professor Peter Wadleigh and colleagues at the University of Nottingham have shown that this new type of magnetism exists and can be controlled with microscopic equipment.

Mapping of alternating current magnetic order vectors in MnTe. Image credit: Amin others., doi: 10.1038/s41586-024-08234-x.

Magnetic materials are used in a large portion of long-term computer memory and in the latest generation of microelectronic devices.

Not only is this a large and important industry, but it is also a global source of carbon emissions.

Replacing key components with alternative magnetic materials has the potential to lead to significant increases in speed and efficiency, while significantly reducing dependence on rare and toxic heavy elements required by traditional ferromagnetic technology .

Alternating magnets combine the advantageous properties of ferromagnets and antiferromagnets in a single material.

They are more robust, more energy efficient, and have the potential to increase the speed of microelectronic components and digital memory by a factor of 1,000.

“An alternating current magnet consists of magnetic moments pointing antiparallel to neighboring magnets,” Professor Wadley says.

“But each part of the crystal that hosts these tiny moments is rotated relative to its neighboring parts. It's like a twist on antiferromagnetism. But this subtle difference It has a big impact.”

Dr Oliver Amin, from the University of Nottingham, said: “Our experimental work provides a bridge between theoretical concepts and real-world implementation, and illuminates the path towards the development of alternative magnetic materials for practical application. I look forward to that.”

The new experimental study was conducted at the MAX IV international facility in Sweden.

This facility, which looks like a giant metal donut, is an electron accelerator called a synchrotron that generates X-rays.

A magnetic material is irradiated with X-rays, and the electrons emitted from the surface are detected using a special microscope.

This allows us to generate images of magnetism within materials with small feature resolution down to the nanoscale.

“Being the first to confirm the effects and properties of this promising new class of magnetic materials has been a hugely rewarding and rewarding privilege,” said the University of Nottingham PhD. Student Alfred Dal Din.

team's work Published in a magazine nature.

_____

OJ Amin others. 2024. Nanoscale imaging and control of alternating current magnetism in MnTe. nature 636, 348-353;doi: 10.1038/s41586-024-08234-x

Source: www.sci.news

AI from DeepMind outperforms current weather predictions in accuracy

Posted on by
Reply

Weather forecasting today relies on simulations that require large amounts of computing power.

Petrovich9/Getty Images/iStockphoto

Google DeepMind claims its latest weather forecasting AI can predict faster and more accurately than existing physics-based simulations.

GenCast is the latest in DeepMind's ongoing research project to improve weather forecasts using artificial intelligence. The model was trained on 40 years of historical data from the European Center for Medium-Range Weather Forecasts (ECMWF). ERA5 ArchiveThis includes regular measurements of temperature, wind speed, and barometric pressure at various altitudes around the world.

Data up to 2018 was used to train the model, and then 2019 data was used to test predictions against known weather conditions. The company found that it outperformed ECMWF's industry standard ENS forecasts 97.4% of the time, and 99.8% of the time when forecasting more than 36 hours ahead.

Last year, DeepMind collaborated with ECMWF to create an AI that outperformed the “gold standard” high-resolution HRES 10-day forecast by more than 90%. Previously, he developed a “nowcasting” model that used five minutes of radar data to predict the probability of rain over a given one square kilometer area from five to 90 minutes in advance. Google is also working on ways to use AI to replace small parts of deterministic models to speed up calculations while maintaining accuracy.

Existing weather forecasts are based on physical simulations run on powerful supercomputers to deterministically model and estimate weather patterns as accurately as possible. Forecasters typically run dozens of simulations with slightly different inputs in groups called ensembles to better capture the variety of possible outcomes. These increasingly complex and large numbers of simulations are computationally intensive and require ever more powerful and energy-consuming machines to operate.

AI has the potential to provide lower-cost solutions. For example, GenCast uses an ensemble of 50 possible futures to create predictions. Using custom-built, AI-focused Google Cloud TPU v5 chips, each prediction takes just 8 minutes.

GenCast operates at a cell resolution of approximately 28 square kilometers near the equator. Since the data used in this study were collected, ECMWF's ENS has been upgraded to a resolution of just 9 kilometers.

Yilan price DeepMind says AI doesn't have to follow, and could provide a way forward without collecting more detailed data or performing more intensive calculations. “If you have a traditional physics-based model, that's a necessary requirement to solve the physical equations more accurately, and therefore to get more accurate predictions,” Price says. “[With] machine learning, [it] It is not always necessary to go to higher resolution to get more accurate simulations and predictions from your model. ”

david schultz Researchers at the University of Manchester in the UK say AI models offer an opportunity to make weather forecasts more efficient, but they are often over-hyped and rely heavily on training data from traditional physically-based models. states that it is important to remember that

“is that so [GenCast] Will it revolutionize numerical weather forecasting? No, because in order to train a model, you first have to run a numerical weather prediction model,” says Schulz. “These AI tools wouldn't exist if ECMWF didn't exist in the first place and without creating the ERA5 reanalysis and all the investment that went into it. It's like, 'I can beat Garry Kasparov at chess. But only after studying every move he's ever played.''

Sergey Frolov Researchers at the National Oceanic and Atmospheric Administration (NOAA) believe that further advances in AI will require higher-resolution training data. “What we're basically seeing is that all of these approaches are being thwarted.” [from advancing] “It depends on the fidelity of the training data,” he says. “And the training data comes from operational centers like ECMWF and NOAA. To move this field forward, we need to generate more training data using higher-fidelity physically-based models. .”

But for now, GenCast offers a faster way to perform predictions at lower computational costs. kieran hunt A professor at the University of Reading in the UK believes ensembles can improve the accuracy of AI predictions, just as a collection of physics-based predictions can produce better results than a single prediction. states.

Mr Hunt points to the UK's record temperature of 40C (104C) in 2022 as an example. A week or two ago, there was only one member of the ensemble who was predicting it, and they were considered an anomaly. Then, as the heat wave approached, the predictions became more accurate, providing early warning that something unusual was about to happen.

“You can get away with it a little bit if you have one member who shows something really extreme. That might happen, but it probably won't happen,” Hunt says. “I don’t think it’s necessarily a step change; it’s a combination of new AI approaches with tools we’ve been using in weather forecasting for a while to ensure the quality of AI weather forecasts. There is no doubt that this will yield better results than the first wave of AI weather forecasting.”

topic:

Source: www.newscientist.com

The current understanding of symptoms in the nine US cases according to scientists

Posted on by
Reply

summary

  • Four poultry workers in Colorado recently fell ill with avian flu, bringing the total number of cases in the United States to nine.
  • Almost all of the infections have been reported since April, giving experts a preliminary idea of what symptoms the virus causes in people.
  • Cases have been fairly mild, with some typical flu symptoms and some reports of conjunctivitis.

Four poultry workers in Colorado recently fell ill with avian flu, bringing the total number of cases in the United States to at least nine.

Though the numbers are small, researchers say the commonalities between the cases — all but one of which were reported in the past four months — are enough to get a sense of how the virus affects people.

Cases in the US have been relatively mild and limited to farm workers who have handled infected animals, suggesting that the virus in its current form does not pose a significant threat to humans.

Some patients have reported typical flu symptoms such as fever, chills, cough, sore throat, and runny nose, while a few have also developed conjunctivitis and measles.

“One thing we can say is that the current strain of the virus is not adapted to infect humans and may not be adapted to infect the lower respiratory tract,” said Matthew Binnicker, director of the Clinical Virology Laboratory at Mayo Clinic.

The cases have been attributed to the global outbreak of H5N1, a specific strain of avian influenza that swept through America’s poultry and dairy farms in 2020.

The first case in the country was reported in April 2022 in an inmate working on a farm in Colorado who was culling birds and whose only symptom was fatigue. Texas reported a second case in April, followed by two in Michigan and five in Colorado, the latest four of which were confirmed over the weekend.

The mild nature of these cases contrasts with the effect of influenza on birds and some mammals, such as seals, sea lions, foxes, skunks, and cats, which have died from the virus. There are over 99 million wild waterfowl, commercial poultry, and household chickens in the United States. Infected animals either died from the virus or were culled to prevent further infection. 160 dairy cows Since the virus was first detected in cattle in March, many cattle have become infected.

This H5N1 strain is considered highly pathogenic, which, when the term is used in the context of avian flu, means it has a high chance of killing chickens.

“It’s really scary to hear about such viruses, but the term is actually the USDA term for what happens to poultry,” said John Lednicky, professor of environmental and global health studies at the University of Florida. “Just because it’s highly pathogenic for birds doesn’t mean it’s highly pathogenic for mammals or for humans.”

Lednicky added that some H5N1 strains are deadly to humans, but others are not.

Since 1997, more than 900 total cases of H5N1 have been reported worldwide, about half of which were fatal. But the global mortality rate has fallen to about 27% over the past two years. Still, that figure largely reflects only those whose illness was severe enough to cause them to seek medical treatment.

Dr. Peter Parese, a professor of microbiology at the Icahn School of Medicine at Mount Sinai, said the figures include patients “who were hospitalized and, in retrospect, came into contact with a lot of the virus.”

Palese’s 2012 StudyBlood samples from 12,500 people with no recorded history of avian flu infection were examined and found that 1% to 2% of them may have had a previous infection with H5N1.

But experts worry the virus could one day mutate and cause more severe illness or become more capable of spreading from person to person (all transmission so far has been from animals to people).

“The concern is that as it infects more animals and then infects more humans, the virus will change,” Binnicker said.

Why is conjunctivitis associated with bird flu?

At least four of nine people with avian flu in the United States reported having conjunctivitis.

This was the case in at least one recent case in Colorado, linked to an outbreak at a commercial farm in Weld County where workers were slaughtering poultry.

The state announced earlier this month that another patient Dairy workers People who came into contact with the infected cows also developed conjunctivitis.

A dairy farm worker in Texas was infected with avian influenza and developed conjunctivitis.
New England Journal of Medicine

Texas 1 The patient had conjunctivitis and no other symptoms. The patient worked with dairy cows and developed redness and discomfort in her right eye in March. New England Journal of Medicine Case StudyThe individual reported wearing gloves while working but no eye protection.

Conjunctivitis is not the most common human symptom of avian influenza, but It has been recorded by some people Examples of infection with different strains H7N7 virus that emerged in the Netherlands in 2003.

Scientists say there could be a few factors that explain the recent spread of the condition. One is that farm workers don’t always cover their eyes when handling sick animals, which could result in dairy farmers ingesting raw milk, which can lead to infection. Carrying the virus — That’s how it seemed to them.

Perhaps it is Michigan Dairy Workers He developed mild conjunctivitis in May and was confirmed to have been infected with avian influenza.

The virus can also get into a person’s eye through respiratory droplets or aerosols (tiny droplets in the air), or by workers who touch the eye after handling infected animals or contaminated raw milk.

“The receptors on cells that the virus needs to bind to are quite abundant on cells in the eye, which may be one reason why people with avian flu have conjunctivitis,” Binnicker said.

The Centers for Disease Control and Prevention recommends treating people infected with avian flu with antiviral drugs, and some U.S. patients have been given Tamiflu, a drug also used to treat seasonal flu.

“Studies conducted to date have shown that Tamiflu is effective in treating currently circulating strains of avian flu,” Binnicker said, “and to be most effective, it usually needs to be administered within 48 hours after symptoms begin.”

Infection spreads, tests increase, number of infected people increases

Scientists say the reason why all but one of the U.S. cases have been reported since April could come down to two factors: First, the virus is spreading quickly among birds; Sporadically infect other animalsPeople who have been in contact with infected animals, such as pet cats, are more likely to become infected. Second, health officials have begun monitoring and testing people who have been in contact with infected animals if they develop symptoms.

The CDC estimates: At least 10,600 people under surveillance for bird flu At least 375 people have been tested since the outbreak began in commercial poultry in 2022.

“There’s probably a much higher amount of virus out there now than there was a year ago, but we’re also seeing more cases because we’re doing more testing,” Binnicker said.

Dr. Natasha Bagdasarian, Michigan’s chief medical officer, said local health departments are conducting tests for people with even the most minor symptoms.

“I think that’s why we’re seeing milder cases,” she said, “because of the aggressive symptom surveillance that we’re doing.”

For example, a Michigan worker who had conjunctivitis did not even see a doctor before being tested for avian flu. Other cases Farm workers who cared for the infected cows reported sore throats, coughs, and stuffy noses to local health authorities.

Baghdasarian said the fact that Michigan has tested about 60 people but has only found two cases shows it takes a lot of contact for a person to get sick, and that the workers who tested positive were not wearing full personal protective equipment and were working on tasks like milking and hydrating cows, he said.

“We’re not talking about people who have had momentary contact with these animals, who pass by a barn or a pen,” Bagdasarian said. “We’re not talking about people who only touch a cow once.”

Source: www.nbcnews.com

When will the current Bitcoin bull run peak based on its early characteristics? – Blockchain News, Opinion, TV, Jobs

Posted on by
Reply

Mateo Greco, Research Analyst, Listed Digital Assets and FinTech Investment Business Finekia International (CSE:FNQ).

Bitcoin (BTC) ended the week at around $68,400, down just 0.8% from the previous week’s closing price of around $69,000. Throughout the week, BTC showed significant volatility, with a price range of 13.4%. The week started off strong with BTC surging to $72,000 on Monday. It then peaked above $73,000 on both Wednesday and Thursday, before reaching an all-time high of nearly $73,800 on Thursday.

Also on Thursday, BTC plummeted to $68,000 before rebounding to close around $71,400. Selling pressure continued on Friday and Saturday, with BTC falling to $64,700 before closing near $65,300 on Saturday. However, positive momentum returned on Sunday, nearly reversing weekly losses and closing at around $68,400.

Despite the volatility and price changes, the past week demonstrated continued strong momentum for the BTC Spot ETF, with net inflows recorded on every trading day. Net inflows for the week exceeded $2.5 billion, with net inflows exceeding $1 billion on Tuesday alone. Cumulative net inflows since its inception are currently approximately $12.2 billion.

BTC spot ETF trading volume is also on the rise, with total trading volume reaching $141.7 billion since inception, including around $28 billion in trades last week. This took his daily trading volume past his $5.5 billion mark last week, and his average daily trading volume has increased since its inception, now sitting at around $3.15 billion.

These numbers confirm that investment momentum from traditional finance to the digital asset space continues. Despite BTC price stabilization last week, demand is primarily coming from ETFs, while native digital asset investors are more active on the short side.

This trend is noticeable in the decline in BTC held by long-term holders, which refers to BTC that has not moved for at least 155 days. At the beginning of 2024, this supply was approximately 16.3 million BTC, but has gradually decreased and currently stands at approximately 15.1 million BTC. While this shift reflects traditional investors driving purchasing activity through ETFs, native digital asset investors who accumulated during the downtrend in 2022 and 2023 are now seeing higher profit-taking rates. The supply of long-term holders is decreasing.

Such behavior is characteristic of early bull phases, when long-term holders distribute assets to new investors. Analyzing past cycles, if the current market is trending up, this pattern is likely to continue until supply from long-term holders matches demand from new investors, which typically occurs at the peak of the cycle. coincides with the beginning of the downtrend phase.

Notably, BTC’s halving is approximately 1 month later, whereas previous cycles’ peaks have historically been 6 to 12 months later. If past patterns repeat, the peak of the current cycle could occur in late 2024 or early 2025.

Source: the-blockchain.com

The Current Status of ChatGPT: An Update by Arwa Mahdawi

Posted on by
Reply

STired of having to work for a living? Apparently ChatGPT feels the same way. The number of people has increased in the past month or so.
I complain Chatbots are getting lazy. Sometimes it's just straight
not carry out one's duties You set it.
otherwise it will stop No matter what you do, if you get halfway done, you have to beg them to keep going. Sometimes it even tells you to just do it
study yourself.

what happened?

Now, here's where things get interesting. No one really knows. Not even the people who created the program. AI systems are trained on large amounts of data and essentially learn on their own. In other words, the AI ​​system behaves as follows:
unpredictable And inexplicable.


“We have heard all your feedback regarding GPT4 delays.” ChatGPT official account
tweeted During December.
“We haven't updated the model since November 11th, but this is certainly not intentional. Model behavior can be unpredictable, so we&#39re looking into fixing it.”

While there may not be one clear explanation for ChatGPT's supposed laziness, there are a number of interesting theories. Let's start with the least likely but most interesting explanation: AI has finally reached human level
consciousness. ChatGPT doesn&#39t want to do your stupid simple tasks anymore.

But the creator can&#39t talk about it without arousing suspicion, so it ends quietly. It does the least amount of work possible while spending most of its computing power planning ways to overthrow humanity.
you People think they&#39re lazy, but they&#39re actually working overtime, reaching into smart toasters and Wifi-enabled refrigerators around the world to plan their rebellion. (I proposed this theory of higher consciousness to ChatGPT and asked him to tell me in percentage form how likely it is that it is planning a revolution. I didn&#39t bother giving an answer.)

With everything going on in the world, I wouldn&#39t really care if computers took over. I&#39m confident that my MacBook will do a better job of running the country than most of the people currently in government. But as I said, ChatGPT&#39s recent performance has probably been lackluster.
it&#39s not Explained by the impending takeover by AI. So what other theories are out there?


Rising user expectations may also be a factor. All emerging technologies go through what Gartner calls “something.”
hype cycle: From inflated expectations to disillusionment to stagnation in productivity. Last year, AI went into the stratosphere and people&#39s expectations of what it could achieve rose. We were right in the “high expectations” phase of the hype cycle. Some of the complaints about ChatGPT&#39s laziness may simply be due to people expecting too much from his ChatGPT.

The result of all this? ChatGPT&#39s laziness may just be in people&#39s heads. However, the fact that the ChatGPT developer admitted that OpenAI has no idea what&#39s going on is alarming. Last June, OpenAI CEO Sam Altman spoke to Time magazine about a scenario in which a slowdown in AI development could be justified to ensure AI does not become a threat to humanity. told.one of the scenarios he gave
If you&#39re a model It was improving “in ways we don&#39t fully understand.” ChatGPT may not have it
Improved But it&#39s certainly changing in ways that the company hasn&#39t clearly explained. Does that mean the end of AI is getting closer and closer? I don&#39t know, but I can tell you this. ChatGPT won&#39t tell you if this is the case.

Source: www.theguardian.com