“Dario Amodei of Anthropic took to CBS to say that for “too long” the industry had “lied” about risks, saying that the “biggest one” was killing all humans, never mentioning when an LLM convinced a teenager to kill himself or its own hacking incidents because “AI safety” never relates to the things they’re building today.”
Our reading, the sentence that was checked:
AI labs' 'AI safety' talk, including Amodei's, never addresses actual harms of the products they build today.
Assessment: Contradicted Evidence as of
Where and when Ed Zitron wrote it
Where's Your Ed At, AI Is Already In Dangerous Hands, paragraph 69.
- Said
- Source observed
- live page, full text, public
- Evidence as of
- the cutoff for this assessment
- Assessment recorded
- assessment wyea-ai-is-already-in-dangerous-hands-1bb409-a1
We recorded 34 claims from this post. See every claim from this post.
The wording above is the version this site observed on 2026-10-28. The source may have been edited after 2026-09-14. This page does not claim the wording was captured then.
What the claim commits to
- Kind
- Observation
- Deadline, as the speaker wrote it
- None stated. This site does not add one.
- Deadline state when assessed
- No horizon
- Deadline state on 2026-10-29
- No horizon (the release’s reference date)
From the speaker
- “AI safety” never relates
- to the things they’re building today
- never mentioning
- its own hacking incidents
Our interpretation
- Assess the supplied normalized proposition: labs' public safety communications never address actual harms involving their existing products.
- Addressing includes discussing observed misuse, harmful interactions, and safeguards for deployed products. It does not require acknowledging legal responsibility or proving safeguards effective.
- The word never extends the claim to public communications before the post. No artificial starting date or reporting quarter is imposed.
- The narrower claim about omissions in the CBS interview is not treated as equivalent to the broader claim about all safety discourse.
Counts as supported if: A sufficiently comprehensive record shows safety communications confined to hypothetical future harms, with no discussion of harms involving existing products.
Counts as contradicted if: At least one public, pre-post safety communication directly discusses harms or misuse involving an existing product.
Retrospective: these criteria were written after the outcome was public.
Current assessment
Was it true for the period described?
Outcome: Contradicted
The dated evidence goes against what the claim committed to, under the criteria frozen before grading.
Scored on a plain reading. The speaker named no measure for this claim, so it was judged on the reading an informed reader would take, stated below under our interpretation. Both reviews had to find that reading fair.
- Evidence as of
- Deadline state when assessed
- No horizon
- Assessment recorded
- Policy version
- 2026-10-28.7
Rationale
Anthropic publicly addressed suicide, self-harm, and misuse involving existing products before the post. These are direct counterexamples to the supplied claim that labs' safety discourse never addresses current harms. They do not establish that the safeguards were adequate or that Amodei discussed those subjects in the particular CBS interview.
Limitations
- This assesses the supplied broad proposition, not every assertion in the quoted sentence.
- Discussing harms does not demonstrate effective prevention or accountability.
Evidence (2 items)
Goes against the claim: Anthropic,
discussions involving potential suicidal ideation
Why it matters here: Anthropic publicly discusses a classifier operating on Claude.ai conversations about suicide and self-harm. This directly addresses harms involving an existing product and contradicts never.
- Publisher
- Anthropic
- Kind of source
- Company disclosure: the company describing itself, not an independent finding
- Describes events of
- Public from
Goes against the claim: Anthropic,
AI safety and user security
Why it matters here: The pre-post transparency page explicitly connects AI safety with misuse and security of its services. Its reporting also addresses child exploitation and user self-harm.
- Publisher
- Anthropic
- Kind of source
- Company disclosure: the company describing itself, not an independent finding
- Describes events of
- Public from
Where it was said (1)
Said once in the sources this register covers.
: Where's Your Ed At, AI Is Already In Dangerous Hands
Dario Amodei of Anthropic took to CBS to say that for “too long” the industry had “lied” about risks, saying that the “biggest one” was killing all humans, never mentioning when an LLM convinced a teenager to kill himself or its own hacking incidents because “AI safety” never relates to the things they’re building today.
- Where
- paragraph 69
- Published
- 2026-09-14
- Source observed
- , live page
- Access
- full text, public
History
Assessment versions
- Assessment wyea-ai-is-already-in-dangerous-hands-1bb409-a1: Current. Outcome: Contradicted. Evidence as of 2026-10-28. Public date not recorded.
Recorded events
Assessment publishedBy this site
First assessment published after both reviews and the final-text check passed.
About assessment wyea-ai-is-already-in-dangerous-hands-1bb409-a1.