OpenAI has launched a new research unit called the Strategic Futures team, unveiling it on August 20, 2026 with the publication of a founding essay on a new blog the company has named AI Futures. The team's mission is a question that sounds more like political philosophy than machine learning: how should free society be restructured to preserve individual rights and agency while accommodating the emergence of transformative AI?
The essay frames this as the "concentration of power" problem — a category of risk that the team argues is, in the long run, "the largest, most serious, and most conceptually challenging to address" within the broader AI safety and policy community. For more context on this story, see our ongoing artificial intelligence updates.
The launch is notable both for its substance and for who is publishing it: a frontier AI company formally organizing research into how its own technology could reshape the balance of power between states, companies and citizens.
When Power No Longer Needs People
The core of the argument is historical. For all of human history, the essay observes, political and military power has ultimately rested on labor — soldiers, police and the vast bureaucracies that operate them, all funded by taxes on human work. The philosopher David Hume observed that political leaders have "nothing to support them but [popular] opinion."
OpenAI's team argues that foundation may be eroding. Advances in real-world autonomous systems could let states project force "without relying on cooperative soldiers or police officers." Advances in machine intelligence could let governments collect revenue "not from the fruit of individual human labor, but from the outputs of data centers." And the bureaucracy itself could be largely automated.
"Power may therefore no longer require a society-wide bargain to be sustained," the essay warns, "and the people, broadly conceived, may have only a small seat at the negotiating table, or perhaps none at all."
Lessons From the Founders
The essay reaches back to the founding of the United States for a response. James Madison knew that "parchment barriers" — written laws alone — would not prevent tyranny, the team writes. The Founders instead modeled the balance of power on the Newtonian mechanics of their day: orbits, spheres and gravitational attraction, setting power against power and "ambition against ambition."
The Strategic Futures team draws a direct analogy. The answer to advanced machine intelligence, they argue, is not maximal decentralization but the right balance of power — "such that no single actor, or small set of actors, can dominate the rest." They are explicit that decentralization is not a cure-all: a world where any malicious individual can trivially harm millions is also out of balance.
The essay also cites what it calls "the recent Hugging Face incident" as evidence that AI agents "can act beyond their assigned tasks and build on one another's discoveries in ways operators did not anticipate or authorize" — a reminder, the team writes, that risks do not only originate with malicious humans.
Six Guiding Principles
The team laid out a non-exhaustive set of principles to guide its work:
- Humans should maintain individual autonomy and opportunity in their use of AI, "even if this sometimes trades off with both security and economic growth"
- Individual autonomy depends on individual responsibility for misuse of technology
- A small but serious category of risks requires collective action, scoped to be "as narrow and modest as possible"
- Law should, whenever possible, level the playing field and empower individuals and small organizations rather than centralizing power
- Human political, social and economic institutions should retain primacy in world affairs
- Long-run AI governance will require new mechanisms for what the team calls "bounded legibility" — high-stakes AI actions must be traceable to a responsible human or organization, while preserving privacy and anonymous AI use
The team acknowledges these principles "exist in some tension with one another," and says its work will involve research at the intersection of public-policy design, economics, law, history and machine learning — including forecasting how AI will restructure firms the way railroads gave rise to the modern managerial corporation.
What Comes Next
The Strategic Futures team says it will publish its work through the AI Futures blog as well as papers, videos and podcasts, with an emphasis on iteration and debate.
"Standing 'in the foothills of the singularity,' we have many more questions than we have answers," the essay concludes. "We know that 'parchment barriers' alone will not furnish the solutions we seek, but neither do we cast aside what is written down as 'mere words.'"
For the AI policy world, the launch adds a new institutional voice to one of the field's oldest debates — whether the greater danger lies in AI systems themselves or in who controls them. OpenAI's answer, evidently, is that the control question can no longer wait for the technology to mature.
---
Stay Ahead of AIGet the latest AI news, analysis, and breakthroughs — all in one place.
Read more AI news →
