AI Transparency Act of 2027: The Omnibus Bill That Won't Fix the Race
AI Transparency Act of 2027: The Omnibus Bill That Won't Fix the Race The AI Transparency Act of 2027 is an omnibus bill that does not fundamentally change the situation.

Automation needs a narrow first win
The best first AI workflow is usually a repeated task with a clear input, clear output, and a human approval step.
AI Transparency Act of 2027: The Omnibus Bill That Won't Fix the Race
The AI Transparency Act of 2027 is an omnibus bill that does not fundamentally change the situation. Congress has recognized the intelligence explosion is near, yet experts warn that on the default path, the next presidential term will see AIs that are far beyond human level.
America now has two workforces: people and AI agents. There are 165 million of them. AI agents are millions of copies spun up and shut down every hour. Most of the work done by these agents is slop, yet people are paying ten billion dollars a month for them. Despite this, no recursive self-improvement has occurred in AI so far.
The bill proposes Plan A as a positive vision: delaying superintelligence until 2040, making research public, and entering mutually assured compute destruction. Humanity should delay the development of superintelligence until 2040, and humanity should begin negotiating something like Plan A as soon as possible. All AI research should be made public, and dozens of companies globally should be allowed to catch up to the frontier of AI.
However, Congress settled that probably not humans will control all these AIs. The discourse bounces back and forth across all options displayed below, and more. Eventually the President and his protégé converge on one plan; the opposition candidate converges on another. By speeding up AI research, the AIs will become even more competent, speeding up research even faster. These AIs will be created entirely by AIs, themselves created entirely by other AIs.
The 2028 election cycle is heated, as usual, with AI being the biggest topic. Both presidential candidates keep getting asked what they'll do about AI. Power is concentrating in the US, and in particular in the President plus a handful of tech CEOs. The President is starting to think about what'll happen to him after he leaves office and the world gets transformed.
Tech industry lobbyists warn that regulation will make America lose the race with China. Other countries are starting to get scared and angry about US and Chinese companies automating white-collar jobs. Most white-collar professions are seeing disruption like software engineering saw in 2026.

Keep reading
Follow the thread
The One-Step Trap: Why Local Accuracy is Killing Long-Term AI Planning
Relying on one step predictions to model the world is a common shortcut in AI research, but it creates a fundamental scaling wall that we can no longer ignore.
Read this noteSame lane, different angle
One-Shot Federated Learning: Trading Weight Averaging for Synthetic Data
Let’s be real: standard federated learning is a communication nightmare.
The Abstraction Gap: Why AI is a Technical Debt Trap
But if you’re actually shipping production code, you know the reality is a lot messier.