Anthropic has a goal of reaching the point where interpretability can reliably detect most model problems by 2027.
Not yet due The date it refers to has not come.
Promise: Commitments to the person's or organisation's own future action, with something to deliver: was it delivered?
Said for an organisation they lead or work for; it counts on both records.
Reasoning
deadline 2027 is after 2026-11-04
Cite this claim
Dario Amodei, “Anthropic has a goal of reaching the point where interpretability can reliably detect most model problems by 2027.” (said 18 February 2025). Verdict: not yet due. Reliability API, claim c_0940ba58bcfd, method 0.1.0+c876030c, dataset release v1.1.2. https://epistacklabs.com/claims/c_0940ba58bcfd
This claim is in dataset release v1.1.2, with the same verdict.
Reply to this claim
More promise claims by Dario Amodei
-
Delivered on time
“We will have web search very soon.”
-
Mostly true
“Anthropic is unilaterally committing to the first of these steps.”
-
Resolved true
“When we receive some kind of formal action, we will look at it, we will understand it, and we will challenge it in court.”
-
Mostly true
“Anthropic is unilaterally committing to this step now.”
Verdicts are made automatically; how verdicts are made.