Advertisement

OpenAI’s $Billions Super App Bet Rests On This One Engineer’s Shoulders

OpenAI is currently carrying out a sweeping overhaul of ChatGPT, with a core goal to reimagine the chatbot’s simple, one-note interface into a personalized AI agent capable of managing tasks across every corner of users’ personal and professional lives. Both internally and publicly, the company has already embraced one name for this ambitious new offering: a “super app.”

This all-in-one platform marks one of the biggest gambles OpenAI has ever made, and the success of that bet now hinges almost entirely on one engineering leader: Thibault Sottiaux. Just last month, Sottiaux was appointed OpenAI’s head of core products, putting him in charge of both ChatGPT and Codex, as well as the project to merge the two tools into the next-generation super app.

To free up focus and resources for the super app, OpenAI has already shuttered multiple standalone projects, including its viral text-to-video tool Sora and an AI platform built for scientific researchers. Most executives leading those shuttered teams have since left the company, while Sottiaux’s influence inside OpenAI has continued to grow. He now reports directly to Greg Brockman, who is overseeing all of OpenAI’s product divisions while Fidji Simo, OpenAI’s CEO of AGI deployment, takes medical leave.

Sottiaux was instrumental in building Codex, which has grown into one of OpenAI’s fastest-expanding revenue streams. Leading Codex previously centered on serving developers and collaborating with AI researchers, but his new role brings an unprecedented new challenge: revamping a consumer product with nearly a billion weekly active users.

“It’s incredibly exciting and mildly terrifying at the same time,” Sottiaux said in an interview earlier this week.

OpenAI has spoken openly and often about its super app plans, but the exact functionality of the final product remains unclear. The term “super app” is most commonly used to describe Asian platforms like WeChat, which bundle everything from messaging to payments and shopping into one interface. But OpenAI is pursuing a far more ambitious vision.

Sottiaux says the end goal is to build the “world’s best personal agent that deeply understands what humans care about.” Over the next year, he says ChatGPT will evolve to become “delightfully proactive,” serving users the right information at exactly the right moment.

Transforming ChatGPT into a super app is OpenAI’s strategy to reignite company growth as it races toward an IPO and fights off stiff competition from Google and Anthropic. OpenAI is betting that a single personalized assistant for every task will reestablish it as the undisputed leader across consumer AI, enterprise AI, and the broader global AI race.

Engineering Roots

Sottiaux grew up in Belgium and studied applied mathematics before joining Google’s London office in 2015. He first worked on Google Maps before moving to Google DeepMind, where he helped build the infrastructure and tools researchers used to develop systems like AlphaGo — the AI that made history in 2016 as the first AI to defeat a human world Go champion.

When ChatGPT launched in 2022, Sottiaux said he felt inspired to relocate to San Francisco and join OpenAI. “This is something that we had been sitting on at DeepMind for almost two years, and we were just not doing it,” he explains of his move.

Sottiaux officially joined OpenAI in 2024, and initially focused on building tools for the company’s in-house researchers, just as he had done at DeepMind. Within a few months, however, he began developing the project that would become Codex. As the AI coding tool exploded in popularity, Sottiaux became a well-known figure in the developer community: he personally responded to bug reports on X, and occasionally granted engineers’ requests to reset their weekly token limits.

But in his new role as head of core products, Sottiaux will now need to prioritize what everyday users want from AI — not just the needs of fellow engineers.

Super App or Super Hype?

In practice, I expect OpenAI’s super app will operate as a digital assistant with advanced memory capabilities. It will likely be able to handle tasks like booking a dinner reservation, then later remind you to skip menu items that trigger your allergies or upset your stomach from your last visit. The platform could also automate work tasks such as filing expense reports before their deadline.

Behind the scenes, Sottiaux says the super app will be powered largely by Codex, which is already seeing strong growth among non-technical users. To complete a task, the agent may write code, run an API call, or browse the web, but users will never see that background work. They will simply make requests in natural language — at least, that is how the system is designed to work.

Sottiaux says building the super app mostly involves converting Codex into a general-purpose agent, then merging that system into ChatGPT. After OpenAI shut down other internal initiatives, the project gained extra resources, though the core team remains relatively small. He declined to share how many people are working on the super app today, but his Codex team only had around 40 members two months ago.

This is not OpenAI’s first attempt to turn ChatGPT into a functional agent. Last year, the company launched Operator, a ChatGPT tool built to navigate the web on a user’s behalf. It eventually morphed into ChatGPT Agent, but neither product ever gained meaningful user adoption. Sottiaux says those attempts were “too early” — the models powering them were not reliable enough, so OpenAI had to heavily restrict what they could do. Now, he claims the technology is ready.

Another flaw in OpenAI’s earlier agents was that consumers did not understand what to do with them. While software developers have mastered using agents to automate a wide range of tasks, teaching everyday users to leverage ChatGPT’s new capabilities will be one of Sottiaux’s biggest challenges.

“We have to bring the user along. Initially, maybe it’s a small thing that we can do for you, and then increasingly, build confidence that ChatGPT can do bigger and bigger things,” says Sottiaux. “Maybe then you start teaching your peers, your friends, and your family these new capabilities that you found in ChatGPT. Then also the model in ChatGPT itself has a role to play there, almost as a mentor.”

Sottiaux would not share a specific launch date for the super app, beyond saying it is coming “soon.” But he notes that “a lot of what is going to be made available for everyone in ChatGPT is already available in the Codex app,” and OpenAI has already confirmed plans to merge Codex into ChatGPT in the coming weeks. Sottiaux adds that OpenAI generally prefers a series of small incremental releases to gather feedback as it goes, in part because the AI space moves so fast that “you can’t really afford to do a big splash and be wrong.”

Not Like the Others

Hundreds of millions of people across Asia have used all-in-one super apps for nearly every online activity for years. OpenAI is pursuing a very different vision, in large part because it has no other choice.

WeChat and Alipay became ubiquitous by building the core financial and information infrastructure that modern China runs on. The U.S., by contrast, already has a well-established ecosystem of tools from Gmail and Instagram to credit cards and Venmo. As a result, OpenAI’s super app will need to integrate with these preexisting systems rather than replacing them.

OpenAI is already moving in this direction. Earlier this week, it announced an expanded partnership with Visa for agentic payments, and it has already built integrations that connect ChatGPT and Codex to users’ email inboxes, Slack, and calendars.

Ultimately, OpenAI is betting it can build a universal consumer interface so powerful and helpful that people will no longer need to think about or interact directly with the underlying websites, apps, and APIs it connects to. But that leaves the company vulnerable to competitors that control the services and infrastructure it relies on. Even so, Sottiaux is convinced we are moving toward a future where everyone will have one single agent to help them navigate their entire life.

“OpenAI is known to take big, bold bets ahead of others, and this is us doing it again,” he says.

This is an edition of Maxwell Zeff’s Model Behavior newsletter. Read previous editions here.

Related Article