Author: TechFlow
Tomorrow, the long-awaited $BIO will officially launch. As a DeSci sector project personally backed by Binance, the market is speculating whether the launch of $BIO will drive a bullish trend in the DeSci sector and siphon off some liquidity from the AI sector.
But are the AI and DeSci sectors necessarily in competition? Not at all. The Solana-based project YesNoError that has been widely discussed recently has taken a path to integrate DeSci and AI, using AI technology to review and discover errors in scientific research papers.
Its token $YNE reached a market cap of $60 million on the day of its launch on December 20th, and was subsequently hyper-promoted by the well-known Twitter KOL Andrew Kang (hereinafter referred to as AK), with a current market cap of around $50 million.
Is it really necessary for AI to review scientific papers?
If you are not yet familiar with the practicality of YesNoError, a clarifying tweet from YesNoError team member Ben Parr explains the necessity of reviewing erroneous information in scientific papers with a real-life example:
In October 2024, a research paper claimed that black plastic kitchenware contained toxins, and this news quickly spread in the media. The Atlantic even published an article titled "Throw Away Your Black Plastic Kitchenware", causing public panic. Even Ben Parr himself started cleaning up his own kitchenware. However, Joe Schwartz, director of the Office of Science and Society at McGill University, discovered an important mathematical error in this study - a simple multiplication error caused the reported toxicity level to be 10 times higher than the actual level. This case shows that even seemingly authoritative research may contain significant errors, and these errors can have a substantial impact on people's lives.
If AI technology is used to review research papers, these basic numerical calculation errors can be avoided to the greatest extent. This is the basis for the creation of YesNoError.
YesNoError was created by Matt Schlicht, using OpenAI's o1 model as the technical foundation. The project's operation is very straightforward: the team uses AI to review research papers and then publicly publishes the issues found on their website yesnoerror.com and official Twitter.
This transparent mode of operation allows the scientific community and the public to be informed in a timely manner of potential issues in important research. Although the project has just started, it has already achieved some notable results, identifying errors in several studies.
The $YNE token has also been endowed with practical use cases, where holders can spend $YNE to have their own papers prioritized for review by the YesNoError AI.
As of now, the YesNoError AI has reviewed 2219 papers and has indeed found a number of errors in them.
Recognition or Skepticism, Voices in the Market
AK is bullish and hyper-promotes
On the day of the $YNE token launch, AK, who has always been bullish on DeSci, expressed his appreciation for the YesNoError project.
AK stated that "the core value of YesNoError lies in the true implementation of cryptocurrency x AI x DeSci."
YesNoError has leveraged the characteristics of the cryptocurrency ecosystem, in which, in this special environment, capital does not require the traditional sense of investment returns. As long as you can attract enough attention, you can get sufficient financial support. (i.e., attention economy, where there are people who pay attention, there will be people who buy the tokens.)
At the same time, YesNoError has also found a good application direction for cryptocurrencies. In the right scenarios, tokens are no longer just hot air, but can actually support public goods that traditional business models cannot sustain.
Perhaps because he is really bullish (or holds a lot of it?), on December 31st, AK published another article introducing and praising the necessity and practicality of YesNoError from a data perspective.
AK stated that YesNoError has the ability to review errors in over 90 million scientific papers in the global literature, which can be completed in just a few weeks or months. If done manually, it would take tens of thousands of years, and even if a team of 5,000 PhDs were assembled, it would still take nearly a decade (and they would not be able to keep up with the pace of new paper publications during that decade), and conservatively estimated to cost $5.4 billion.
However, with the optimized AI model, it only takes about $30 million ($ 0.3 per paper) to complete the review work more accurately and in a more standardized manner - less than 1% of the cost of the manual method.
In the traditional scientific field, raising $30 million is also a considerable undertaking, but in the crypto space, it is much easier. (Although it contains many speculative factors, the market cap of $YNE has already reached $50 million in just ten days.)
Currently, this AI agent has reviewed more than 1,700 papers and found an error rate of around 3-4%. And going forward, its processing speed will be further improved through continuous optimization. Among the 90 million papers, there may be many important papers containing significant errors, and correcting these errors could have a substantive positive impact on the world.
The official account of the BIO Protocol also agrees with AK's view:
Is it a false demand? Let's look at different voices
In addition to the bullish voices, there are also those who question the real demand for YesNoError.
Kyle Samani, co-founder of Multicoin Capital, expressed opposition in the comments of AK's bullish article:
Kyle believes that according to the 80/20 rule, only a small number of papers are truly important, and these important papers, being under sufficient scrutiny, are unlikely to have known errors.
However, Andrew Kang refuted this with data. He pointed out that even by Kyle's logic, among the 90 million papers, assuming only 5% are important, that still means there are 4.5 million important papers. Even if the error rate in these important papers is only 0.1%, it still means there are 4,500 important papers with errors that need to be corrected. And the "black shovel research" case mentioned earlier fully demonstrates that even papers with major impact can contain errors and have a certain impact on society.
Summary
The idea of using AI to review papers is not entirely new, as there have been many AI-based paper review use cases since the emergence of ChatGPT. From the perspective of the crypto space, the emergence of YesNoError may not only solve the problem of scientific paper errors, but also have some real-world applications for cryptocurrencies beyond just speculation (of course, it is still in the early stages of the project, and some of its value is still dependent on the market's speculative enthusiasm).
As for the market behavior, although much of the bullish behavior in the market can be summarized as "the ass decides the brain", if the project is truly viable and has practical and useful value beyond speculative hype, this "making money while standing" behavior is likely to be recognized by the market as well.
How YesNoError will develop in the future remains to be seen, and we need to see if the project team maintains the determination to keep going after the market hype subsides.
We hope to see more projects that truly benefit the world.
