OpenAI Is Reportedly Readying ‘o,’ an AI Agent to Answer Meta’s Muse
Trending Topics Jakob Steinschaden ● Covered by 5 sources
Rumor — unconfirmed reporting.
OpenAI may show “o” at DevDay: a personal AI agent that keeps working after you close the chat. It’s a direct answer to Meta’s Muse, and the race is now about trust, access and who gets there first.
Based on reporting by Trending Topics, Jakob Steinschaden — read the original for the full story.
Summary, retelling and take written by AI under human oversight; images are AI-generated illustrations. How we work · Report an error
OpenAI looks ready to answer Meta on its own turf. At DevDay in San Francisco on Tuesday, the company could introduce a personal AI agent called “o” that keeps running in the background and works around the clock. Nothing is official yet, but the hints have been piling up fast.
The strongest clues came from TestingCatalog. The leak tracker says ChatGPT’s configuration includes multiple references to “o,” including a display name and an email suffix ending in “-o,” which points to a possible email identity for the agent. For a short time, “o” also showed up on the Pro upgrade page, where the $100-a-month plan was described as an “always-on assistant” that keeps going after the chat window is closed.
There’s also an internal name floating around: Aeon, OpenAI’s label for continuously running agents. Reports, including one from Wccftech, expect “o” to use a variant of the GPT-6 Astra model tuned for long tasks. Rumors say it may be available across all Pro tiers, while a new ChatGPT Pro Max plan at around $500 a month is also said to be coming with especially fast responses. The details are still loose, including permissions, launch timing in Europe and even the final name.
That launch would fit the shape of OpenAI’s wider push. The company has already introduced continuously running Workspace Agents for business customers, so a consumer version would not come out of nowhere. Peter Steinberger, the Austrian developer behind OpenClaw, joined OpenAI earlier this year, and Sam Altman said at the time that he would drive the next generation of personal agents. But there is no confirmation that he worked on “o” or will appear at DevDay.
The pressure is coming from Meta. Muse launched about three weeks ago for a mass audience in the U.S., and Meta says it can connect to email and calendars, write emails, book travel, fill out forms and complete purchases. It runs after the app is closed, asks for approval when needed, and has already climbed to the top of the free iPhone apps in the U.S. while passing 2.5 million downloads in two weeks, according to Sensor Tower figures cited by CNN. OpenAI is clearly not waiting to be the second act.
My take — AI-written commentary, not fact-checked reporting
This is the new agent war: who gets a software helper trusted with your inbox, calendar and payments before the other guys do. Meta is sprinting, OpenAI is reacting, and everyone else is pretending this is just product iteration. It isn’t; it’s a land grab for the most intimate layer of software, with privacy concerns politely waiting at the door.
Read more about this at: Trending Topics
