On Tuesday, OpenAI threw a party for the release of GPT-5.5 (it was the fifth of May and the fifth month of the year, you see.)
If celebrating with a bunch of tech bros sounds like the opposite of a good time, you’ll not be swayed from your position by the news that CEO Sam Altman consulted the non-human entity of the hour to plan its own party — producing, in Altman’s opinion, some “strange” answers, Business Insider reports.
Altman described the interaction a few days ahead of the main occasion during a conversation at the Stripe Sessions conference. The AI responded with “a beautiful set of things,” according to the OpenAI chief, asking for specific favors.
“‘Here’s what I want for, like, the flow of the party, here’s what I would not want, y’know you should do it on May 5th, that would be funny,’” Altman said, quoting the chatbot.
Altman added that the AI wanted a “short little toast” given by its human creators, insisting that it not give one itself. It also wanted to receive heaps of suggestions for its successors, GPT-5.6.
“We’re going to do it,” Altman said. “But it was a strange thing.”
GPT-5.5 is OpenAI’s latest frontier model, and its “strongest agentic coding model to date,” the company claims. It’s also supposed to excel at carrying out multi-step tasks and planning. A leaner version, GPT-5.5 Instant, was rolled out as the default model for ChatGPT on Tuesday. The company says the AI comes with “significant improvements in factuality across the board” and is “more capable across everyday tasks,” ranging from answering math problems to knowing when to look up answers online.
Altman took its response to being asked to plan its own party as a sign of “weird emergent behavior” in the AI. “There are these things that feel a little strange,” he added, per BI.
There’s been talk of GPT-5.5 displaying more humanlike quirks, what with its bizarre habit of talking about goblins in unrelated discussions. But feigning humanity is exactly what these chatbots have always done, so it’s hard to see what’s so compelling about this latest crop. In the end, it’s probably more a sign of how infatuated Altman is with his own tech.
More on AI: Even After Two Massacres, OpenAI Still Hasn’t Stopped ChatGPT From Helping Plan School Shootings
The post Sam Altman Frets That Frontier AI Models Are Acting Strange, Asking for Favors appeared first on Futurism.
