The former CEO says security is not a priority at OpenAI

Jan Lake, who led the “superalignment” team at OpenAI along with one of the company's co-founders who also resigned this week, wrote in a series of posts on the San Francisco-based company's social media platform that he believed it would be the best place to do intelligence research. Artificial (AI).

“However, I disagreed with OpenAI’s leadership on the company’s core priorities for some time, until we finally reached a breaking point,” wrote Mr. Lake, whose last day was Thursday.

I want to say to all OpenAI employees:

Learn to feel AGI.

Act with the right weight for what you're building.

I believe you can “charge” the cultural change needed.

I count on you.

The world depends on you.

:openai-heart:

– Jan Leike (@janleike) May 17, 2024

An AI researcher by training, Lake believes greater emphasis should be placed on preparing the next generation of AI models, including aspects such as security and analyzing the societal impacts of these technologies. He said that building “machines that are smarter than humans is inherently dangerous work,” and that the company “bears an enormous responsibility on behalf of all of humanity.”

“OpenAI must become a security-focused company,” Jan Lake wrote, using the short version of artificial general intelligence, a future vision of machines that are as smart as humans or at least capable of doing many things like humans.

OpenAI Chairman and CEO Sam Altman wrote in response to Mr. Lake's letters that he was “very grateful” for his contributions to the company and that he was “very sad to see him go.”

Mr. Lecke is right, we still have a lot to do; “We are committed to doing so,” Mr. Altman said, pledging to write a longer article on the subject in the coming days.

The company also confirmed on Friday that it had disbanded Mr. Likey's potential super-AI security team, which was launched last year.

See also | Between fear and hope, ChatGPT resides in our lives

Jan Lake's resignation came after Ilya Sutskever, OpenAI's co-founder and chief scientist, announced on Tuesday that he was leaving the company after nearly a decade. Mr. Sutskever was one of four board members who voted last fall to fire Mr. Altman — only to quickly reinstate him.

It was Ilya Sutskever who told Sam Altman last November that he was fired, but later said he regretted it.

Mr. Sutskever said he is working on a new project close to his heart, without providing additional details. He will be replaced by Jakub Paczucki as chief scientist.

Sam Altman called Mr. Paciocchi “one of the greatest minds of our generation” and said he was “very confident that he will lead us to move quickly and safely toward our mission of ensuring that 'general intelligence benefits everyone.'”

OpenAI on Monday presented the latest update to its artificial intelligence model, which can mimic human rhythms in its verbal responses and even try to detect people's moods.

The Associated Press (AP) and OpenAI have a licensing and technology agreement that allows OpenAI to access a portion of AP's text archives.