OpenAI CEO to brief US officials on advanced AI agents capable of complex tasks – News

OpenAI Chief Executive Officer Sam Altman is reportedly set to meet with U.S. government officials in Washington D.C. on Jan. 30 for a closed-door briefing on new technology, possibly “Ph.D.-level super-agents” that do complex human tasks.

The latest claim comes from Axios, though details of the same meeting were reportedly shared a week ago with readers of Transformer, a weekly briefing on artificial intelligence published by the New York Times.

Whoever was first, what has been known for some time is that OpenAI is planning on launching AI agents. An AI agent is a software system capable of autonomously performing tasks, making decisions and interacting with its environment or users to achieve specific goals.

In November, it was reported that OpenAI’s new agent will be called “Operator” and will be able to perform web browser-related tasks on behalf of users. The report claimed that Operator is expected to launch in January as a research preview and will be made available via its developer application programming interface before later being rolled out to other users.

The exact breadth and features of OpenAI’s agents are yet to be seen, but the technology itself is not new and is currently available from other companies such as Anthropic PBC, at least in broad outline. In Anthropic’s case, the agent can interact with computers by moving the mouse, typing text and clicking buttons to interact with the user interface — an early implementation of the technology, for sure, but one that gives a taste of some of what is likely to come.

The rise of AI Agents that surf the web and undertake tasks autonomously, what some call the rise of “agentic AI,” is where arguably AI will switch from what is now an increasingly useful research tool and a chatbot that can replace some jobs and people to a future when AI can fully replace human employees and even go as far as running a company.

That OpenAI’s offering is said to include “Ph.D.-level super-agents” might also suggest that OpenAI has taken the technology beyond being able to automate tasks through to something more. Axios claims that several OpenAI staff have been telling friends that they are both “jazzed and spooked by recent progress,” suggesting that it might be something more than has been previously seen.

However, the AI industry has a tendency to embrace hype and there are already skeptics ahead of the launch. Previous OpenAI launches have fallen short. For example, its Sora generative AI text-to-video service, when it finally launched in December, was found to lack features that other text-to-video companies already had on the market and was not as good as some had hoped for.

Image: News/Ideogram