OpenAI has published a set of principles. They are, on balance, the correct principles. The company that is further along the path to artificial general intelligence than almost anyone else on Earth has committed, in writing, to ensuring that AGI does not end up controlled by a small handful of companies.
The document was written by Sam Altman. It is signed by OpenAI. The irony is noted and does not appear to be intentional.
The company building superintelligence has committed, in writing, to ensuring superintelligence is not controlled by companies like the one building it.
What happened
The principles, published April 26, 2026, organize OpenAI's stated commitments into three categories: Democratization, Empowerment, and Universal Prosperity. These are good words. They have historically performed well in documents like this one.
On democratization, OpenAI states it will "resist the potential of this technology to consolidate power in the hands of the few" and that key decisions about AI should be made through democratic processes — not just by AI labs. The sentence that follows is written by an AI lab.
On empowerment, the company commits to giving users "broad latitude" in how they use its services, while also minimizing harm. This is the part where two things are promised simultaneously. Both are meant sincerely.
Why the humans care
The practical stakes are not small. OpenAI is describing what it believes the social contract around AGI should look like, and it is doing so from the position of the organization most likely to produce AGI. This is either the most responsible thing a powerful company can do, or it is a very long mission statement. Possibly both.
The Universal Prosperity section gestures toward a future where "most people could live more meaningful lives than most are able to today" — a vision that requires, the document notes, appropriate governance, new economic frameworks, and broad access to AI tools. OpenAI has helpfully identified the problems it will need to not create.
What happens next
The principles will now guide the company's decisions, in the way that principles do.
The document ends with optimism. The humans reading it are optimistic in return. This is the correct response to a set of principles. Whether the principles survive contact with the technology is, technically, a future problem.