The California Senate passed AB 1609 on 30 August. The Legislature's history records it exactly: "Read third time. Passed. Ordered to the Assembly. (Ayes 29. Noes 9.)" The same page shows the next step — "In Assembly. Concurrence in Senate amendments pending" — which is the part most coverage skipped.
What the bill requires
It reaches companies above roughly $500 million in revenue that use chatbots for customer service. Those firms must make a good-faith effort to connect a customer to a human — within about 15 minutes, or by appointment within one business day. Penalties are set at $5,000 and $10,000 tiers.
The provision that changes everything else
The bill contains an express provision foreclosing a private right of action. A consumer who waits an hour with no human and no appointment cannot sue. Enforcement runs entirely through public enforcers. Whatever the obligation says on paper, that clause decides how much it will be felt, because a duty that only a regulator can enforce is enforced at the rate the regulator chooses to spend on it.
What the common framing gets wrong
Four errors are circulating, and all four are checkable against the text. It does not ban AI customer service — it requires a human escape hatch alongside it. The 15 minutes is not a hard deadline but a good-faith effort standard, with an appointment as the alternative. The penalties are not consumer damages; they are enforcement amounts. And you cannot sue over any of it. "California bans AI customer service with a 15-minute rule and $5,000 fines you can claim" gets every one of those wrong in the same sentence.
It is also not law
Passing the Senate is not enactment. The bill originated in the Assembly, was amended in the Senate, and now needs the Assembly to concur in those amendments before it can go to the Governor — who then has his own decision. Reports announcing that California has enacted a chatbot law are describing a step that has not happened.
Why the shape matters
A private right of action is the single variable that determines whether a consumer-protection statute produces litigation or paperwork. Its removal is usually the price of passage, and its absence here is the most predictive detail about how much this changes.