Tag Archives: doubt

I Highly Doubt the Existence of a “Nutrition Electron Microscope”

Posted on by
Reply

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.

A new kind of microscope?

Science is one of the most fruitful sources of new terms. There are conditions such as “mitochondrial integration” and “quantum fluctuations” and there is no way to make sentences reliable.

Recently, there have been various scientific papers that contain the phrase “.”Nutrition Electron Microscope/Microscope“The term suggests a device for scanning broccoli, but it is completely nonsense. There are scanning electron microscopes and tunneling electron microscopes, but there are no nutritional electron microscopes.

One possible explanation was proposed by Alexander Magazinov, a software engineer who illuminates Moonlight as a watchdog for Science Publishing. He pointed to Article from 1959 in Bacteriological reviewthe text was formatted into two columns. 4 To the bottom of the pagethe words “nutrition” and “electron microscope” appear next to each other in the left and right rows. Older papers are often scanned using optical character recognition, but such software can be a pain to deal with complex formats. “Nutritional electron microscope“According to the magazine, it is “artificial for text processing.”

But the journalist on Retraction Watch I discovered another possibilitythat was it Reddit has been flagged. In Falsi, the phrases “scanning electron microscope” and “nutritional electron microscope” are very similar, and, importantly, they use almost identical letters. The only difference is a single dot, nuqta. This means that small mistakes in translating paper from Persian to English are sufficient to create a “nutritional electron microscope.”

These explanations are not mutually exclusive, and feedback is satisfied that they can explain the appearance of this phrase. The bigger question is why it lasts in published research. Are these papers not strict? Peer reviews and checksto ensure high accuracy and therefore maintain the integrity of the scientific literature? Perhaps such “tortured phrases” should be included in the checklist of warning signs that the paper may be plagiarized or fraudulent.

Readers who encounter similar tortured phrases during their viewing of technical literature are invited to submit them to their regular address.

The nun is too far away

Sometimes feedback can receive stories that feel so good. The setup is so clean and the rewards are amazingly inevitable at the same time, so we doubt ourselves. Is the reality very beautiful? And we remember that the Titanic faction was the largest ship ever on that maiden voyage when it was built and when bad things happened. Sometimes reality is melodramatic. So, I believe this story happened as explained, but it may not be.

Come to us from Charlie Watnaby. The late Father John, Charlie Watnaby, was a curator at the Science Museum in London. It is inevitably related to the issue of Scunthorpe. The difficulty of banning offensive words in online discussions when strings of the same letters can appear in harmless words such as “peacock” or “sussex.”

John’s story is, technically speaking, not an example of Scunthorpe’s problem, but it definitely contiguous to it. As Charlie explains, “On the early days of the Computing Gallery, machines were set up so that the public could enter their own words and see them on the big screen.

This may seem like an invitation to misconduct. Therefore, readers will be pleased to know that staff expect an inevitable attempt to write a torrent of filth on a big screen so that everyone can see. They drew a “long list of blasphems,” all of which were blocked.

“Everything was going well,” says Charlie, until the system was defeated by the most dangerous person possible: the computer expert. While trying to use the machine, he realized that some keystrokes did nothing. “After investigating, he was able to pull up the entire list of offence (or offensive) words on the big screen so that everyone could see.

Feedback is prepared to believe in 90% of this story, but in the absence of independent verification, it draws a line to the nun. But we are willing to do wrong about this too. If the abbey schoolchild was at the science museum on that fateful day, and if you think you remember, contact us.

Yodel-eh-oh

Senior news editor Sophie Bushwick has turned his attention to a press release entitled “.Monkeys are the best in the world Yoderer – New Research.” It describes research examining the “special anatomy” of the throats of apes and monkeys, known as vocal membranes. These membranes allow for “the same rapid transition of frequency heard in alpine yodering” but “a much more praised range”, sometimes “over three musical octaves.”

After such accumulation, there was a breathless feedback accompanied by feedback and feedback was made to find it Audio Recording A tufted cappuchin monkey. We were hoping for the diffusive appeal that sparked. Music sounds Or the focus of the Dutch rock yodeler. What we got was “Skroark Rark Eek.” And now we understand why Sophie said, “I can’t stop laughing.”

However, if you look closely, you will notice the missed opportunities. Do not hesitate to show us the “yodering” of the tufted cappuchin. However, this study also included Howler Monkey.

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.

Source: www.newscientist.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

The Big 7 tech companies are questioning the potential of the AI boom – What’s driving the doubt? | Artificial Intelligence (AI)

Posted on by
Reply

It’s been a tough week for the Grand St. Seven, a group of technology stocks that have played a leading role in the U.S. stock market, buoyed by investor excitement about breakthroughs in artificial intelligence.

Last year, Microsoft, Amazon, Apple, chipmaker Nvidia, Google parent Alphabet, Facebook owner Meta and Elon Musk’s Tesla accounted for half of the S&P 500’s gains. But doubts about returns on AI investments, mixed quarterly earnings, investor attention shifting elsewhere and weak U.S. economic data have hurt the group over the past month.

Things came to a head this week when the shares of the seven companies entered a correction, with their combined share prices now down more than 10% from their peak on July 10.

Here we answer some questions about Seven and the AI boom.

Why did AI stocks fall?

First, there are concerns that the huge investments being made by Microsoft, Google and others in AI will pay off. These have been growing in recent months. Goldman Sachs analysts The memo was published In June, the Wall Street bank released a report titled “Gen AI: Too Much Spending, Too Little Reward?” which asked whether $1 trillion in investment in AI over the next few years “will ever pay off,” while an analysis by Sequoia Capital, an early investor in ChatGPT developer OpenAI, estimated that tech companies would need $600 billion in rewards to recoup their AI investments.

Gino said “The Magnificent Seven” is also hit by these concerns.

“There are clearly concerns about the return on the AI investments that they’re making,” he said, adding that big tech companies have “done a good job explaining” their AI strategies, at least in their most recent financial results.

Another factor at play is investor hope that the Federal Reserve, the U.S. central bank, may cut interest rates as soon as next month. The prospect of lower borrowing costs has boosted investors’ support for companies that could benefit, such as small businesses, banks and real estate companies. This is an example of “sector rotation,” in which investors move money between different parts of the stock market.

Concerns about the Big 7 are affecting the S&P 500, given that a small number of tech stocks make up much of the index’s value.

“Given the growing concentration of this group within U.S. stocks, this will have broader implications,” said Henry Allen, macro strategist at Deutsche Bank AG.Concerns about a weakening U.S. economy also hit global stock markets on Friday.

What happened to tech stocks this week?

As of Friday morning, the seven stocks were down 11.8% from last month’s record highs, but had been dipping in and out of correction territory — a drop of 10% or more from a recent high — in recent weeks amid growing doubts.

Quarterly earnings this week were mixed. Microsoft’s cloud-computing division, which plays a key role in helping companies train and run AI models, reported weaker-than-expected growth. Amazon, the other cloud-computing giant, also disappointed, as growth in its cloud business was offset by increased spending on AI-related infrastructure like data centers and chips.

But shares of Meta, the owner of advertising-dependent Facebook and Instagram, rose on Thursday as the company’s strong revenue growth offset promises of heavy investment in AI. Apple’s sales also beat expectations on Thursday.

“Expectations for the so-called ‘great seven’ group have perhaps become too high,” Dan Coatsworth, an analyst at investment platform AJ Bell, said in a note this week. “These companies’ success puts them out of reach in the eyes of investors, and any shortfall in greatness leaves them open to harsh criticism.”

A general perception that tech stocks may be overvalued is also playing a role: “Valuations have reached 20-year highs and they needed to come down and take a pause to digest some of the gains of the past 18 months,” says Angelo Gino, a technology analyst at CFRA Research.

The Financial Times reported on Friday that hedge fund Elliott Management said in a note to investors that AI is “overvalued” and that Nvidia, which has been a big beneficiary of the AI boom, is in a “bubble.”

Can we expect to see further advances in AI over the next 12 months?

Further breakthroughs are almost certain, which may reassure investors. The biggest players in the field have a clear roadmap, with the next generation of frontier models already underway to train, and new records are being set almost every month. Last week, Alphabet Inc.’s Google DeepMind announced that its system had set a new record at the International Mathematical Olympiad, a high school-level math competition. The announcement has observers wondering whether the company will be able to tackle long-unsolved problems in the near future.

The question for labs is whether these breakthroughs will generate enough revenue to cover the rapidly growing costs of achieving them: The cost of training cutting-edge AI has increased tenfold every year since the AI boom really began, raising questions about how even well-funded companies such as OpenAI, the Microsoft-backed startup behind ChatGPT, will cover those costs in the long run.

Is generative AI already benefiting the companies that use it?

In many companies, the most successful uses of generative AI (the term for AI tools that can create plausible text, voice, and images from simple prompts) have come from the bottom up: people who have effectively used tools like Microsoft’s Copilot or Anthropic’s Claude to figure out how to work more efficiently, or even eliminate time-consuming tasks from their day entirely. But at the enterprise level, clear success stories are few and far between. Whereas Nvidia got rich selling shovels in the gold rush, the best story from an AI user is Klarna, the buy now, pay later company, which announced in February that its OpenAI-powered assistant can: Resolved two-thirds of customer service requests In the first month.

Dario Maisto, a senior analyst at Forrester, said a lack of economically beneficial uses for generative AI is hindering investment.

“The challenge remains to translate this technology into real, tangible economic benefits,” he said.

Source: www.theguardian.com