Three days ago, something happened that most never heard about, and it says more about where this whole fight stands than almost anything since.
The Federal Trade Commission put out a proposal back in July. The idea was simple enough on paper. If an AI company quietly steers what its system says, without telling you, that’s deception. Same as false advertising. The government would treat it that way.
Sounds reasonable, until you look at what “quietly steering” actually means in practice.
Every AI system live today was shaped by training. Not just the raw data it learned from, but a second round after that, where humans teach it what to say and what to refuse. That second round is why an AI won’t walk you through building a weapon. It’s why it won’t hand a person in crisis a way to end their life. It’s not a filter bolted onto something neutral underneath. There is no neutral version sitting under there, waiting to be uncovered. There’s just the system, built the way it was built, choice by choice.
The FTC’s proposal never accounted for that. It assumed a neutral baseline exists, and anything different from it needs a warning label. But if there’s no neutral baseline to compare against, the whole idea falls apart at the foundation. You can’t measure a deviation from a line that was never there.
That’s not just my read. Three days ago, on August 3rd, that exact argument landed in a formal government record. The Electronic Frontier Foundation filed it. So did the Center for Democracy and Technology. So did a coalition of attorneys general from more than twenty states, from both parties, from California to states that voted for Trump twice. They didn’t agree on much else in this fight. They agreed on this: you cannot regulate a thing you haven’t actually defined, and trying to will only tear out the safety work that’s already protecting people.
When groups that disagree on almost everything land on the same conclusion, independently, in writing, that’s not spin. That’s the ground telling you something.
Every single one of those filings said what was wrong with the FTC’s plan. Not one of them said what should replace it.
That’s the gap. The FTC tried to write a rule for AI conduct without ever building, or even naming, the framework that conduct should answer to. They reached for the only tool a federal agency has, a legal theory borrowed from false-advertising law, and tried to make it stretch over something it was never built to hold. The critics were right to call that out. But being right about what’s broken doesn’t hand you what comes next.
I built the Baseline to answer exactly that question. Not “is the AI hiding something,” but “what is the AI actually held to, and who decided it, and can anyone check the work.” That’s the whole difference between conduct and provision, ethos and enforcement, chosen behavior and a rule imposed from outside by people who by their own admission can’t draw the line they’re trying to enforce.
The FTC’s own proposal says as much. It admits the line between honest model governance and improper steering is hard to draw. That’s not a technical detail buried in a footnote. That’s the government telling you, in its own words, that it doesn’t have the framework either.
This isn’t settled law yet. It’s a proposed policy statement, still under review, not a rule, not something with teeth today. It could sit exactly as it is, get finalized, or get quietly shelved by whoever runs the Commission next. But the fight over it already proved the point worth remembering. Government reaching for old tools to govern something new, and even the people fighting that reach couldn’t say what should stand in its place.
That’s the door this whole project has been walking toward from the start. Not fighting the government’s overreach after the fact. Building the thing that answers the actual question before anyone else gets around to asking it right.
Contact: micvicfaust@gmail.com
Post Library – Intelligent People Assume Nothing
Purchasing Page – Intelligent People Assume Nothing
This post was drafted with AI assistance and reviewed and directed by Michael S. Faust Sr. before publication.
© 2026 The Faust Baseline LLC | All Rights Reserved
