Archive note: Backfilled on June 28, 2026 as part of the Florida Desk rebuild; publication date is set to the weekly archive slot. Claims were checked against the sources listed below.
The federal AI debate keeps asking what Congress should write. Florida is asking what existing law can already reach.
That is the heart of the AI liability gap. There is no comprehensive national AI statute. There are, however, consumer-protection laws, negligence theories, product-safety arguments, youth-safety duties, public-nuisance claims, and discovery rules.
My Read
I think this is where the next real AI governance record gets built. Not in a hearing room first, but in subpoenas, complaints, discovery fights, and internal safety files.
That is uncomfortable, but it is also predictable. If a company markets a system to the public, especially to or around minors, the question cannot be only what the model said in one bad moment. It has to include what the company knew before launch, what it tested, what it warned, and what it chose to emphasize in public.
State attorneys general can move faster than Congress. That power can be abused, but it can also force sunlight where voluntary transparency has failed.
Why I Care
The liability gap will not remain a gap. If lawmakers do not fill it with statutes, courts and attorneys general will fill it with claims, settlements, and records. Companies should assume the next AI governance fight is not a panel. It is a subpoena.
What I Am Watching
- Whether states focus first on minors and vulnerable users.
- Whether safety-evaluation records become discoverable.
- Whether labs change marketing language before courts decide the merits.