Anthropic executive Cat Wu joins OpenAI to lead core products
In a major move within the artificial intelligence industry, Cat Wu has officially joined OpenAI to lead product management for its Core Products. The transition comes just two days after she announced her exit from rival AI firm Anthropic on November 3, 2026.
Taking to X on November 5, Wu shared her excitement about her new role. She noted her deep respect for OpenAI, praising the company for proving that scaling works, bringing AI tools to over a billion people, and making great products central to its mission.
In her post, Wu wrote: "I’m excited to work with @thsottiaux, an incredible PM team, and everyone across OpenAI. Together, we’ll build agents that have your back and your team’s, and keep you in control, so you get more time for the people and work you love. Lots to learn, and lots to build. Please keep your feedback coming!"
She received a public welcome from OpenAI President and co-founder Greg Brockman, who was recently highlighted in the company's October 21 Google Cloud announcement. Thibault Sottiaux, known as @thsottiaux on X, also welcomed her, saying he looked forward to working together. Sottiaux currently serves as OpenAI's Core Products Lead.
Wu brings deep hands-on experience in building AI tools. She spent two years shaping Claude Code at Anthropic, joining when it was just a small internal project. During her time there, she famously asked the AI to build a reinforcement-learning environment and guided it for dozens of hours without writing code manually.
Upon leaving, she credited Claude Code, Cowork, and the teams, along with the users whose feedback shaped their work.The Future of Persistent AI Agents
Wu's goal of keeping users in control lines up perfectly with OpenAI's upcoming plans. In June 2026, OpenAI announced its planned acquisition of Ona, a startup specialising in secure cloud execution. The goal is to integrate Ona's technology into OpenAI's Codex, which already boasts over 5 million weekly users, a 400 per cent jump from earlier in the year.
Currently, Codex has expanded beyond software development into research, analysis, and automation. The next step is creating "persistent agents." This means the AI can keep working on tasks for hours, even after you close your laptop, operating securely inside a customer's own cloud environment.
Users will be able to check progress, give directions, make decisions, and review the final results remotely.
For businesses, this level of control is crucial. Organisations need strict boundaries regarding where agents run, what data they can access, and how their work is logged and reviewed.
By combining Wu’s experience in delegating large tasks to AI with OpenAI’s persistent cloud models, the company aims to solve a big challenge: expanding what AI can do on its own while ensuring humans can still direct, inspect, and govern it easily.- EndsAlso Read | AI startup hacks Anthropic in 12 hours, gains access to employee details and internal messagesAlso Read | Anthropic warns it is nearing Self-Building AI, what it means for the futureAlso Read | OpenAI launches LifeVal to test if AI can actually handle everyday tasks
I am a journalist covering the ever-evolving consumer tech landscape. From smartphones to laptops, I track the latest drops and decode the data to see if the newest gadgets are worth the hype.
