“They more power we give them (but hooking up them to agents, people, etc), the more risk we are incurring.”
Our reading, the sentence that was checked:
Giving LLMs more power, such as hooking them up to agents, increases risk.
Assessment: Can’t be checked Evidence as of
Where and when Gary Marcus wrote it
Marcus on AI, Humanity’s “Oh shit!” AI moment?, paragraph 7.
- Said
- Source observed
- live page, full text, public
- Evidence as of
- the cutoff for this assessment
- Assessment recorded
- assessment marcus-humanitys-oh-shit-ai-moment-46f81e-a2
We recorded 10 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 2024-12-12. This page does not claim the wording was captured then.
What the claim commits to
- Kind
- Causal claim
- 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)
Conditions
- giving LLMs more power, e.g. via agents (qualifier, stated by the speaker)
From the speaker
- the more power we give them
- hooking them up to agents, people, etc
- the more risk we are incurring
Our interpretation
- This is a general claim that risk rises with LLM power or autonomy. It names no measure of risk, no baseline, and no harm threshold.
- 'Risk' here means exposure to possible future harm. That is a judgment about likelihood. A single observed outcome cannot confirm or refute it.
- Reports of harm from agentic AI show that harm happened. They do not measure whether risk grows with the amount of power granted. So no plain test of the claim exists.
This claim is not graded on evidence. The record checks that the words are quoted correctly and attributed to the right person, and that the claim is classified correctly.
Current assessment
Not graded on evidence
Outcome: Can’t be checked
The statement is a recommendation, a value judgment, or too open to test. It is recorded, not scored.
- Evidence as of
- Deadline state when assessed
- No horizon
- Assessment recorded
- Policy version
- 2026-10-28.7
Rationale
The claim is that the more power LLMs are given, for example by connecting them to agents or people, the more risk we take on. It names no measure of risk, no baseline, and no scale for power. Anthropic's threat report, published 27 August 2025 (e1), says agentic AI was used to carry out cyberattacks. That shows harm from agentic use. It does not compare harm across different levels of permission, autonomy, or access. The record has no controlled comparison of that kind. Because the claim's terms are open-ended and the record has no matching test, the claim cannot be scored here.
Limitations
- The only evidence item is one company's report on misuse of its own models. It records incidents and measures nothing about how harm scales with the power granted.
- The record does not show whether harm stays flat or falls when LLMs get more permissions. It neither supports nor contradicts the claim's causal form.
Evidence (1 item)
No outcome evidence: a claim of this kind is recorded and attributed, not scored.
Where it was said (1)
Said once in the sources this register covers.
: Marcus on AI, Humanity’s “Oh shit!” AI moment?
They more power we give them (but hooking up them to agents, people, etc), the more risk we are incurring.
- Where
- paragraph 7
- Published
- 2024-12-12
- Source observed
- , live page
- Access
- full text, public
History
Assessment versions
- Assessment marcus-humanitys-oh-shit-ai-moment-46f81e-a2: Current. Outcome: Can’t be checked. 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 marcus-humanitys-oh-shit-ai-moment-46f81e-a2.