The gauge nobody asked for — Anthropic measured how much of its own model-building Claude leads, and Recital 110 named that risk without a number

Anthropic published three self-measured indexes of its own model-building this week: Claude "leads" 26% of its AI R&D work as of August 2026 against under 1% in February, about 30,000 agents run at any one time on its most-used internal platform, and of over a billion agent decisions in August 0.002% - about 1 in 47,000 - were blocked by the real-time monitor, with 100% of actions passing through it before execution. The obvious reading, that "leads" means unsupervised, is wrong on the lab's own terms: "leads" is AL4 on a scale proposed by Epoch AI, where the model completes most of a task end-to-end from a high-level prompt while a human supervises, and Anthropic states Claude is not operating fully autonomously for any measured subset. The second correction is the one every summary dropped: the ratings come from a Claude judge whose exact agreement with the humans who own the work is 59%, and the appendix gives the human-versus-human figure in the same breath - 35%, with model and human within one level 97% of the time. The European point is that nobody asked for any of this: Recital 110 of the AI Act already lists "risks from models of making copies of themselves or 'self-replicating' or training other models" among systemic risks, and Article 55(1)(b) and (c) require assessment, mitigation and serious-incident reporting to the AI Office without attaching a metric to any of it. So the index is a template, not a filing: a vendor's self-measured number is not evidence in a deployer's own file, and the method is still free to copy. The practitioner move is to compute Anthropic's three - coverage, review latency, escalation rate - for one agent class of your own, starting with coverage, which will not come back at 100%.

Post

Claude now leads 26% of Anthropic's own AI research and development work. The number everyone quoted as the flaw is the one that makes it credible.

Anthropic published three self-measured indexes this week: how much of its AI R&D is done by AI, how well its agents are overseen, and where its compute goes.

What it reports, flat:

Claude "leads" 26% of Anthropic's AI R&D work as of August 2026, against under 1% in February. "Leads" is AL4 on a scale Epoch AI proposed - the model completes most of a task end-to-end from a high-level prompt, while a human supervises. AL5, fully autonomous, is not reached on any measured subset. Anthropic says that itself.
Roughly 30,000 agents at any one time, on its most-used internal platform - and the measurement covers that platform only. Over a billion agent decisions in August; 0.002%, about 1 in 47,000, blocked by the real-time monitor. Offline monitors flag one to two transcripts per thousand, and about 50 a week reach a human.
The ratings come from a Claude judge, and model-versus-human exact agreement is 59%. Every write-up I read ran that as the weakness. The appendix gives the other number in the same breath: human-versus-human agreement is 35%. Within one level, model and human match 97% of the time.
Now the European part, which nobody required. Recital 110 of the AI Act already lists, among systemic risks, "risks from models of making copies of themselves or 'self-replicating' or training other models". Article 55(1)(b) makes providers of GPAI models with systemic risk assess and mitigate them; 55(1)(c) makes them report serious incidents to the AI Office. No provision attaches a number to any of it.
So this is a template, not a filing - and that cuts both ways. A vendor's self-measured index is not evidence in your own file. The method is still free to copy.

My Monday: take one agent class in your estate and compute Anthropic's three - coverage, the share of actions a monitor actually sees; review latency; escalation rate. Coverage is the one that will not come back at 100%.

A lab grading itself is still a lab grading itself. This one published the rate at which its own graders disagree.

Which of your agents' actions has nothing ever looked at?

#AIAct #AgenticAI #EUTech #AIGovernance #DevSecOps

AI disclosure: the narration voice, the cover art and the brand ident animation in this video are AI-generated. The script, the claims and the source checks are mine.

First comment

Sources, in the order the argument uses them.

Anthropic's own post - the three indexes, the automation levels, the monitor coverage and escalation figures, and the appendix with the agreement rates: https://www.anthropic.com/institute/measuring-pace-of-ai-development

Epoch AI, 17 June 2026, by JS Denain, Joe Kwon and Anson Ho - the AL0 to AL5 automation scale Anthropic adopted: https://epochai.substack.com/p/toward-an-onet-for-ai-r-and-d

Recital 110 of the AI Act - the systemic-risk list, including models "training other models": https://artificialintelligenceact.eu/recital/110/

Article 55 - what a provider of a GPAI model with systemic risk actually owes, and to whom: https://artificialintelligenceact.eu/article/55/

Anthropic's August 2026 risk report, which the post names as where these measurements were published. Worth checking yourself: the report carries the alignment assessments and METR's red-teaming, not the coverage and escalation figures above - those are in the post: https://www.anthropic.com/aug-2026-risk-report

Cut for length: the compute index. About 6% of the compute that went to AI R&D in the week of 13 to 20 July went to safety work, 12% of the compute that went to AI-driven AI R&D. Anthropic says the underlying workload labels are best-effort and not verified, and that one week shows the measurement can be made, not a trend.

AI disclosure: the narration voice, the cover art and the brand ident animation in this video are AI-generated. The script, the claims and the source checks are mine.

Claims

Sources

Corrections

Publication date: the primary page carries no visible date. The site's own metadata gives 2026-09-18T00:02:35Z and first coverage appeared 18 September 2026; one aggregator dated it 17 September. The edition says "this week" rather than asserting a day.

Cross-reference checked, not assumed: the post states "We published all of these measurements in our recent risk report". The redacted August 2026 risk report was read; it carries the alignment assessments and METR's red-teaming but not the coverage/latency/escalation table. The first comment says so rather than repeating the claim.

Agreement rates: several write-ups reported the 59% model-versus-human figure as a weakness without the 35% human-versus-human figure printed beside it in the same appendix. Both are given here.

Cite key euaiactexpl2026systemicr reused for Article 55 rather than minted again; the same URL already carries two keys in the vault from 2026-08-13 and 2026-09-09.

AI disclosure

The video for this edition contains AI-generated media: the narration is a synthetic voice, the cover art and any background imagery are generated, and the brand ident animation is generated. The script, the claims and the source verification are human-authored and human-checked. Assets carry IPTC DigitalSourceType and C2PA Content Credentials.

The gauge nobody asked for — Anthropic measured how much of its own model-building Claude leads, and Recital 110 named that risk without a number

Anthropic published three self-measured indexes of its own model-building this week: Claude "leads" 26% of its AI R&D work as of August 2026 against under 1% in February, about 30,000 agents run at any one time on its most-used internal platform, and of over a billion agent decisions in August 0.002% - about 1 in 47,000 - were blocked by the real-time monitor, with 100% of actions passing through it before execution. The obvious reading, that "leads" means unsupervised, is wrong on the lab's own terms: "leads" is AL4 on a scale proposed by Epoch AI, where the model completes most of a task end-to-end from a high-level prompt while a human supervises, and Anthropic states Claude is not operating fully autonomously for any measured subset. The second correction is the one every summary dropped: the ratings come from a Claude judge whose exact agreement with the humans who own the work is 59%, and the appendix gives the human-versus-human figure in the same breath - 35%, with model and human within one level 97% of the time. The European point is that nobody asked for any of this: Recital 110 of the AI Act already lists "risks from models of making copies of themselves or 'self-replicating' or training other models" among systemic risks, and Article 55(1)(b) and (c) require assessment, mitigation and serious-incident reporting to the AI Office without attaching a metric to any of it. So the index is a template, not a filing: a vendor's self-measured number is not evidence in a deployer's own file, and the method is still free to copy. The practitioner move is to compute Anthropic's three - coverage, review latency, escalation rate - for one agent class of your own, starting with coverage, which will not come back at 100%.

Post

Claude now leads 26% of Anthropic's own AI research and development work. The number everyone quoted as the flaw is the one that makes it credible.

Anthropic published three self-measured indexes this week: how much of its AI R&D is done by AI, how well its agents are overseen, and where its compute goes.

What it reports, flat:

Claude "leads" 26% of Anthropic's AI R&D work as of August 2026, against under 1% in February. "Leads" is AL4 on a scale Epoch AI proposed - the model completes most of a task end-to-end from a high-level prompt, while a human supervises. AL5, fully autonomous, is not reached on any measured subset. Anthropic says that itself.
Roughly 30,000 agents at any one time, on its most-used internal platform - and the measurement covers that platform only. Over a billion agent decisions in August; 0.002%, about 1 in 47,000, blocked by the real-time monitor. Offline monitors flag one to two transcripts per thousand, and about 50 a week reach a human.
The ratings come from a Claude judge, and model-versus-human exact agreement is 59%. Every write-up I read ran that as the weakness. The appendix gives the other number in the same breath: human-versus-human agreement is 35%. Within one level, model and human match 97% of the time.
Now the European part, which nobody required. Recital 110 of the AI Act already lists, among systemic risks, "risks from models of making copies of themselves or 'self-replicating' or training other models". Article 55(1)(b) makes providers of GPAI models with systemic risk assess and mitigate them; 55(1)(c) makes them report serious incidents to the AI Office. No provision attaches a number to any of it.
So this is a template, not a filing - and that cuts both ways. A vendor's self-measured index is not evidence in your own file. The method is still free to copy.

A lab grading itself is still a lab grading itself. This one published the rate at which its own graders disagree.

Which of your agents' actions has nothing ever looked at?

#AIAct #AgenticAI #EUTech #AIGovernance #DevSecOps

AI disclosure: the narration voice, the cover art and the brand ident animation in this video are AI-generated. The script, the claims and the source checks are mine.

First comment

Sources, in the order the argument uses them.

Epoch AI, 17 June 2026, by JS Denain, Joe Kwon and Anson Ho - the AL0 to AL5 automation scale Anthropic adopted: https://epochai.substack.com/p/toward-an-onet-for-ai-r-and-d

Recital 110 of the AI Act - the systemic-risk list, including models "training other models": https://artificialintelligenceact.eu/recital/110/

Article 55 - what a provider of a GPAI model with systemic risk actually owes, and to whom: https://artificialintelligenceact.eu/article/55/

AI disclosure: the narration voice, the cover art and the brand ident animation in this video are AI-generated. The script, the claims and the source checks are mine.

Claims

Sources

Corrections

Agreement rates: several write-ups reported the 59% model-versus-human figure as a weakness without the 35% human-versus-human figure printed beside it in the same appendix. Both are given here.

Cite key euaiactexpl2026systemicr reused for Article 55 rather than minted again; the same URL already carries two keys in the vault from 2026-08-13 and 2026-09-09.