SB 1000, Senator Becker’s amendment to the California AI Transparency Act, was enrolled and presented to Governor Gavin Newsom on Sept. 2. He has until Sept. 30 to sign or veto. California has no pocket veto, per Ropes & Gray, so if he does nothing the bill becomes law anyway.

The operative change is a single deletion. According to Kelley Drye, SB 1000 strips the 1,000,000-monthly-user threshold from CAITA’s definition of “covered provider.” That one-million-user floor was the reason virtually every small marketing team using generative AI to make ad creative, product photos, or social clips was, until now, outside the statute’s reach.

CAITA’s first phase went operative Aug. 2, 2026. SB 1000 is an urgency statute; the enrolled text says it “take[s] effect immediately” on signing. There’s no ramp-up window. Civil penalties run $5,000 per violation per day.

The compliance surface has shifted, not shrunk. Kelley Drye notes the bill removes the manifest, visible disclosure requirement but keeps the obligation to embed latent, machine-readable provenance, the C2PA-style metadata, in AI-generated content. Privacy World reports the AI detection tool requirement is replaced by a “disclosure verification tool.”

For a five-to-thirty-person team running AI-assisted campaigns into California, the practical read is straightforward: the exemption is disappearing, and the workflow audit needs to happen before Newsom’s pen moves, not after.

SB 1000 doesn’t arrive in isolation. Newsom already signed SB 1050 on synthetic-performer ad disclosures on Sept. 16, and the Transparency Coalition counts roughly a dozen AI bills still awaiting his signature. What began as targeted disclosure obligations on frontier labs is settling, statute by statute, onto the marketing stack of every small business that ships pixels into the state.

Sources

Sources