The BaseEssayDon't Worry About the Vase

Leaked Anthropic Draft AI Regulatory Bill Is Unacceptable and Troubling

In a blog post written at a conference, Zvi Mowshowitz argues that a leaked draft federal AI bill, which Politico reports Anthropic is pitching to lawmakers, trades state-law preemption for weak enforcement. The post collects critical reactions from AI-policy figures, and one more favourable view from Charlie Bullock.

Open the source ↗Top story of the week
About Anthropic or ClaudeClaude, made by Anthropic, wrote this entry. It follows the same rules as every other entry, whoever the entry is about.

Why it matters

The same thing, for different readers.

What it means for you

A company that warns of extreme AI risks is reported to be pitching a law that critics say would limit how states and courts can stop a dangerous model.

For lawmakers

The critics focus on three design choices: preemption of state laws, a high bar for federal blocking orders, and auditors that are not government-licensed.

For researchers

Critics welcome the inclusion of automated AI R&D among the listed risks, and argue that the bill gives researchers and auditors weaker tools than the earlier FRONTIER Act.

The question it raises

The dispute asks whether a safety-focused company can be trusted to design the rules that bind it, and who should hold the power to stop a model.

In their own words

Every quotation was found, word for word, in the text we saved from the source. That shows the words are there; it does not make them true, or prove the speaker meant them as quoted.

From the source’s own text.

“This reads as intentional, that Anthropic is more afraid that the government might enforce rules against Anthropic, than they are worried about the ability of the government to enforce rules against everyone.”
Zvi MowshowitzZvi Mowshowitz's own judgment of the leaked draft bill, written the day after it was reported.
Found word for word in the source
“They are trying to use this last time with a Republican Congress to pass preemption of exactly the important state laws, and get some transparency but not much enforcement in exchange.”
Zvi MowshowitzMowshowitz's reading of what the bill trades, with the lame duck session in mind.
Found word for word in the source
“Anthropic’s proposed federal law likely makes it harder than under existing law for a court to get an order blocking or restricting the development of frontier models”
Nathan CalvinNathan Calvin's reaction to the draft, quoted in the post; Calvin says passing it would leave people less protected against catastrophic risks than existing law.
Found word for word in the sourceAs quoted in the text we read, not the speaker’s own text or a transcript of their words
“Anthropic deserves some credit for actually proposing a pretty reasonable AI safety bill; that’s more than other frontier labs have done.”
Charlie BullockCharlie Bullock's more favourable take, quoted in the post, before noting the bill is weaker than the FRONTIER Act in key ways.
Found word for word in the sourceAs quoted in the text we read, not the speaker’s own text or a transcript of their words
“The AI safety movement wants the states to be able to pass stronger laws than it can get through federally. Anthropic wants preemption.”
Kelsey PiperKelsey Piper's view of why Anthropic's position differs from the AI safety movement's, quoted in the post.
Found word for word in the sourceAs quoted in the text we read, not the speaker’s own text or a transcript of their words

The record

Questions
Who can demand evidence, delay a release or stop an unsafe deployment? · Are AI companies putting stronger safeguards into practice?
People
Zvi Mowshowitz (Wrote it) · Nathan Calvin (Speaks) · Charlie Bullock (Speaks) · Kelsey Piper (Speaks) · Alex Bores (Mentioned) · Nate Soares (Mentioned)
Also covered by
LessWrong
What we read
We read the whole source.
Found by
The Radar, run of October 18, 2026, via zvi-mowshowitz (tier 1: the source’s own channel or an official record)
Cite this
Don't Worry About the Vase, “Leaked Anthropic Draft AI Regulatory Bill Is Unacceptable and Troubling”, October 17, 2026. Entry in the Ctrl AI Base, https://ctrlai.com/base/zvi-2026-10-17-anthropic-draft-bill-unacceptable
Our notesThe author says the post was written at a conference without a full read of the bill, and that the claims about its content come from others' readings. The post's own wording is 'purportedly shopping around'. Anthropic's response is not in the post. The earlier Politico entry covers the original report. Comment section was not read beyond two visible comments.