Anthropic Founders Publicly Needle OpenAI Execs

Estimated read time 2 min read


Image for article titled Anthropic Founders Publicly Needle OpenAI Execs

Photo: Maxwell Zeff

Dario and Daniela Amodei, siblings and founders of Anthropic, fired indirect jabs at OpenAI while not stating the company’s name on Thursday. The founders told a San Francisco audience that Anthropic’s AI was the best in the world, and defended their competitive advantage over industry leader, OpenAI.

“Claude 3 Opus is the most capable and powerful AI model available anywhere in the world,” said Daniela Amodei at Thursday’s Bloomberg Technology Summit.

The Amodeis, former OpenAI employees who defected to create a safety-focused AI company, chose their words carefully to never mention their former company, but made several remarks that indirectly referenced the ChatGPT maker. Sam Altman has been on an interview tour lately noting how OpenAI’s latest AI model, GPT-4 “kinda sucks” compared to what’s coming. However, the Anthropic founders were not shy to get behind their large language models.

“We have 7 cofounders,” said Dario Amodei. “Three and a half years later, we’re all still at the company.”

The remark seemed to jab at OpenAI, which has lost several founding members in the last few months. Founding member Andrej Karpathy left OpenAI in February. Co-founder Ilya Sutskever has not been heard from in months, and OpenAI’s management has dodged questions about his employment since November when he reportedly was part of the effort to fire Sam Altman.

Bloomberg BusinessWeek’s Brad Stone peppered Anthropic’s founders with questions about the Federal Trade Commission’s investigation into AI startup’s relationships with cloud providers. Anthropic’s billion-dollar partnerships with Google and Amazon are part of that. However, the Amodeis reassured the audience that Anthropic’s partnership with two different cloud providers makes them different from “some of those other deals.” Presumably, this meant OpenAI which has a tight partnership with Microsoft.

At other times, Anthropic’s founders issued a word of caution about AI. The makers of Claude noted that AI could learn from the failures of the social media world. Daniela Amodei said the move fast and break things mentality that defined Silicon Valley’s social media era did do quite a bit of harm. Amodei said it’s important to approach this next phase of technology carefully.



Source link

You May Also Like

More From Author

+ There are no comments

Add yours