Openai and FDA are talking about the use of artificial intelligence in drug evaluation

Food and The drug administration was meeting with Openai To discuss the agency’s use of Amnesty International, according to sources with the knowledge of the meetings. Meetings seem to be part of a wider effort in the Food and Drug Administration to use this technology to accelerate the process of approval of the drug.

“Why does it take more than 10 years until a new property comes to the market?” Books of the Food and Drug Administration Marti Macari on X Wednesday. “Why did we not verify artificial intelligence and other things? We just completed the first scientific review with the help of AI for a product, and this is just a start.”

The notes came after an annual meeting of the American Hospital Association earlier this week, as Makari talked about the capabilities of artificial intelligence to help the approval of new treatments for diabetes and certain types of cancer.

Makary did not specify that Openai was part of this initiative. But the sources close to the project say that a small team of Openai met with the FDA and two of his colleagues for the so -called government efficiency management Elon Musk several times in recent weeks. The group discussed a project called CderGPT, which is likely to be Narcotics Evaluation CenterWhich organizes medications that do not need prescription in the United States, and the GPT research. Jeremy Walsh, who was recently named as the first officer of artificial intelligence in the FDA (FDA), led the discussions. To date, no contract has been signed.

Openai refused to comment.

Walsh also met with Peter Bowman Davis, a university student on Leaves from Yale University who is currently head of the AI ​​team in the Ministry of Health and Humanitarian Services, to discuss AI’s aspirations in the Food and Drug Administration (FDA). Politico I mentioned for the first time the appointment of Bowman Davis, who is part From Andrink Horwatz American dynamic team.

When it was reached via e -mail on Wednesday, Robert Calif, who held the position of FDA from 2016 to 2017 and again from 2022 to January, said that the agency’s review teams have been using artificial intelligence several years ago. “It will be interesting to hear the details of the review parts that were” with the help of Amnesty International “and what does that mean.” “There has always been seeking to shorten the review times and widespread consensus that artificial intelligence could help him.”

Before Calif left the agency, he said that the Food and Drug Administration (FDA) is considering the different ways in which Amnesty International can be used in internal operations. “The final reviews of approval are only one part of a much greater chance,” he says.

In order to be clear, the use of artificial intelligence to help with final drug reviews represents an opportunity to pressure a small part of the timeline to develop long drugs. The vast majority of medications fail before it reaches a review of the Food and Drug Administration.

“He supports automation of some tasks related to the drug review process, but he says that there should be directives to politics about the type of data used to train AI models and the type of performance that the model considers is acceptable. Says, but he says that there should be directives to politics about the type of data used to train AI models and the type of performance that the model considers is acceptable. He says: “These machines are incredibly ingenious in learning information, but they must somehow, so they learn what we want to learn,” he says.