Academics accuse the startups of artificial intelligence to review the subscriber of propaganda

There is a clouds controversy on “AI’s created” studies to the ICLR for this year, a long -term academic conference that focuses on artificial intelligence.

At least three Amnesty International Labs – roadblockand biologyAnd Car science – The claim that it used artificial intelligence to create studies that have been accepted in ICLR workshops. At conferences like ICLR, the workshop organizers usually review studies for publication at the conference workshop.

Sakana informed ICLR leaders before submitting her papers created from artificial intelligence and obtained the approval of peer auditors. ICLR spokesman confirmed on Techcrunch.

Many academics have moved from artificial intelligence to social media to criticize autonomous science and autocience repair as a participation in the process of reviewing scientific peers.

“All of these stormy cards of artificial intelligence use places that peers reviewed like human Evals, but no one agreed to provide this free work,” wrote Berthfiraj Amanabolo, Assistant Professor of Computer Science at the University of California, San Diego. X post. “This makes me lose the respect of all concerned, regardless of how impressive is.

As critics noticed, a long -time peer review is a long time, the density of work, most of which are volunteer. According to the survey of modern nature40 % of academics spend two to four hours reviewing one study. This work was escalating. The number of papers provided to the largest artificial intelligence conference, NeuPs, increased to 17,491 last year, an increase of 41 % of 12345 in 2023.

Academic circles have already faced a problem of copies of artificial intelligence. One analysis Find Between 6.5 % and 16.9 % of the papers presented to artificial intelligence conferences in 2023 probably contained artificial text. But artificial intelligence companies that use the pendant review to measure their technology and announce it effectively are relatively new.

(Technology) received positive review papers unanimously, “he wrote archaeology in a After x ICLR results. In the same post, the company continued to claim that workshop auditors praised one of the “smart ideas” created by AI.

Academics did not look gently on this.

Ashwinee Panda, post -PhD fellow at the University of Maryland, He said in X Publishing The presentation of papers created by artificial intelligence without giving the organizers of the workshop the right to reject them showed “the lack of respect for the time of human auditors.”

Panda added: “Sakana continued to ask us if we were ready to participate in their experience in the workshop that I organize in ICLR,” Panda added, “I (we) no (…) I think that the introduction of artificial intelligence papers to a place without contacting the auditors.”

Not for nothing, many researchers are skeptical that the papers created of artificial intelligence deserve the visionary effort.

Sakanana herself I confess Artificial intelligence has committed “embarrassing” martyrdom errors, and that only one of the three papers created from the artificial intelligence that the company chose to provide will meet the tape to accept the conference. The company said that Sakana pulled the ICLR sheet before it was published from the Exploration of Transparency and respecting the ICLR agreement.

Alexander Doria, co -founder of AI Startup Pleias, said that a group of pure artificial offers that indicate the need for a “high -quality study” general company/public agency that was created from artificial intelligence at a price.

Doria in a series From jobs On X. “Academic circles are not present to use an external sources (AI) Evals.”