Team update
OpenAI
OpenAI just said its team has grown to 45 people. Small crew, big ambitions: they're working on everything from software systems to robots.
Based on reporting by OpenAI — 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 dropped a short update this week, and the headline number is 45. That's the size of the team right now, according to the company itself, spread across work that ranges from testing out new research ideas to building software infrastructure to putting machine learning onto actual robots.
There's not a lot of detail beyond that. No breakdown of roles, no mention of funding, no specific projects named. Just a statement that the group is pushing on what the post calls the frontier of AI capabilities, across three fairly different lanes: validating novel ideas, creating new software systems, and deploying ML on robots.
What's notable is the shape of that list. It's not just researchers publishing papers. It's engineers building systems and people getting machine learning models to actually run on physical hardware. That's a broader spread of work than you'd guess from a pure research lab, and it hints at ambitions that go past theoretical benchmarks into things that move and act in the real world.
Forty-five people is tiny by today's standards, where AI labs routinely employ thousands. But this update reads like an early snapshot of a company betting that a small, focused team can cover research, software, and robotics all at once, rather than picking one lane and staying in it.
My take — AI-written commentary, not fact-checked reporting
I like reading these old-school update posts because they show how unglamorous the early days actually were, no funding rounds, no drama, just a headcount and a to-do list. It's a useful reminder that scale doesn't equal progress, and that the labs obsessing over billion-parameter models today started as a few dozen people fiddling with robots in a room.
Read more about this at: OpenAI