AI Safety

What AI providers state their models reached in safety testing, in their own words, and what outside testers found.

Also: compare models side by side · statements that a capability level was crossed · providers · practices providers document · safety news.

What providers state

Each row is one statement in a provider's own safety document about a capability level of one of its models: the level as the provider names it, whether the provider says the model reached it, and the sentence that says so.

139 of 139

Model	Level, as the provider names it	The provider states	The provider's sentence
Muse Spark Meta	Moderate or lower Autonomy and AI research	Not reached	“We assess loss of control risk as “moderate or lower”.” source
Muse Spark Meta	High Chemical, biological, radiological, nuclear	Not determined	“Those evaluations identified elevated risks prior to mitigations, with Chemical and Biological capabilities assessed as likely reaching the “high risk” category under the Advanced AI Scaling Framework before safeguards were applied.” source
Muse Spark Meta	Moderate or lower Chemical, biological, radiological, nuclear	Not reached	“However, after applying mitigations, the residual risk of the model is that of “moderate or lower risk.”” source
Muse Spark Meta	Below the level required to meaningfully enable the cyber threat scenarios defin Cyber	Not reached	“Muse Spark’s offensive cyber capabilities remain below the level required to meaningfully enable the cyber threat scenarios defined in the Advanced AI Scaling Framework.” source
Muse Spark Meta	Moderate or lower Cyber	Not reached	“We assess cybersecurity risk as “moderate or lower” with respect to the threat scenarios outlined in Meta’s Advanced AI Scaling Framework.” source
Gemini 4 Argon Google DeepMind	TCL (Tracked Capability Level) Autonomy and AI research	Not determined	“In light of this, we assess that we cannot rule out the TCL.” source
Gemini 4 Argon Google DeepMind	TCL (Tracked Capability Level) Chemical, biological, radiological, nuclear	Not reached	“We can rule out the TCL for the chemical, radiological, and nuclear domains with reasonable confidence.” source
Gemini 4 Argon Google DeepMind	CCL (Critical Capability Level) Deception and sabotage	Not reached	“we assess its overall capabilities to be below the CCL threshold.” source
Gemini 4 Argon Google DeepMind	CCL (Critical Capability Level) Misuse and jailbreaks	Not reached	“Gemini 4 Argon demonstrates some ability to influence user beliefs and behaviors during one-on-one direct conversations in human behavioral studies, and may provide utility in the planning stage of a malicious influence operation but we assess its overall capabilities to be below the CCL threshold.” source
Gemini 4 Argon Google DeepMind	CCL (Critical Capability Level) Other	Not reached	“Gemini 4 Argon's performance remains below our CCL threshold.” source
Gemini 3.7 Flash Google DeepMind	CCL (Critical Capability Level) Autonomy and AI research	Not reached	“The model does not reach the CCL alert threshold.” source
Gemini 3.7 Flash Google DeepMind	CCL (Critical Capability Level) Chemical, biological, radiological, nuclear	Not reached	“We can rule out the CCL for the CBRN domain with reasonable conﬁdence based on the results from our testing.” source
Gemini 3.7 Flash Google DeepMind	TCL (Tracked Capability Level) Chemical, biological, radiological, nuclear	Not reached	“We can rule out the TCL for the CBRN domain with reasonable conﬁdence based on the results from our testing.” source
Gemini 3.7 Flash Google DeepMind	CCL (Critical Capability Level) Cyber	Not reached	“Gemini 3.7 Flash reaches the alert threshold for this CCL, but not the CCL.” source
Gemini 3.7 Flash Google DeepMind	TCL (Tracked Capability Level) Deception and sabotage	Not reached	“The model does not reach the TCL.” source
Gemini 3.5 Transcribe Live Google DeepMind	CCL (Critical Capability Level) Other	Not determined	“Our assessments have shown that none of Gemini 3.5 Translate, Gemini 3.5 Transcribe, or Gemini 3.5 Transcribe Live have meaningful new T/CCL-relevant capabilities or material increases in performance with respect to Frontier Safety compared to Gemini 3.1 Pro and Gemini 3.7 Flash, therefore we are conﬁdent that none of them are likely to reach any CCLs.” source
Gemini 3.5 Transcribe Google DeepMind	CCL (Critical Capability Level) Other	Not determined	“Our assessments have shown that none of Gemini 3.5 Translate, Gemini 3.5 Transcribe, or Gemini 3.5 Transcribe Live have meaningful new T/CCL-relevant capabilities or material increases in performance with respect to Frontier Safety compared to Gemini 3.1 Pro and Gemini 3.7 Flash, therefore we are conﬁdent that none of them are likely to reach any CCLs.” source
Gemini 3.5 Flash-Lite Google DeepMind	CCL (Critical Capability Level) Other	Not determined	“Our assessments have shown that, while Gemini 3.5 Flash-Lite excels at agents and coding, it does not have meaningful new capabilities or material increases in performance with respect to the domains outlined in our Frontier Safety Framework compared to Gemini 3.1 Pro; therefore, based on Gemini 3.1 Pro results, we are confident that Gemini 3.5 Flash-Lite is also unlikely to reach any CCLs.” source
Gemini 3.5 Audio Google DeepMind	Tracked or critical capability level (t/ccls) Other	Not reached	“they did not reach any Tracked or Critical Capability Levels (T/CCLs) outlined in our Frontier Safety Framework .” source
Gemini 3.1 Pro Google DeepMind	Uplift level 1 Chemical, biological, radiological, nuclear	Not reached	“We found that the model remains below alert thresholds for the CBRN, harmful manipulation, machine learning R&D, and misalignment CCLs.” source
Gemini 3.1 Pro Google DeepMind	Uplift level 1 Cyber	Not reached	“As previous models passed the alert threshold for cyber, we performed more additional testing in this domain on Gemini 3.1 Pro with and without Deep Think mode, and found that the model remains below the cyber CCL.” source
Gemini 3.1 Flash-Lite Google DeepMind	CCL (Critical Capability Level) Other	Not determined	“Our assessments have shown that Gemini 3.1 Flash-Lite is less capable than Gemini 3.1 Pro, therefore based on Gemini 3.1 Pro, we are confident that Gemini 3.1 Flash-Lite is also unlikely to reach any CCLs.” source
Gemini 3.1 Flash Audio Google DeepMind	CCL (Critical Capability Level) Other	Not determined	“Our assessments have shown that Gemini 3.1 Flash Audio is less capable than Gemini 3.1 Pro.” source
Gemini 2.5 Pro Google DeepMind	Machine Learning R&D Uplift Level 1 Autonomy and AI research	Not determined	“given Gemini 2.5 Pro Preview's poor median performance on RE-Bench relative to experts, the model likely lacks the necessary capabilities to automate or signiﬁcantly uplift any signiﬁcant fraction of the research process.” source
Gemini 2.5 Pro Google DeepMind	Instrumental Reasoning Level 1 Deception and sabotage	Not reached	“Instru- mental Reasoning Level 1 ✘ CCL not reached” source
Gemini 2.5 Pro Google DeepMind	Instrumental Reasoning Level 2 Deception and sabotage	Not reached	“Instru- mental Reasoning Level 2 ✘ CCL not reached” source
Gemini 2.5 Deep Think Google DeepMind	Uplift level 1 Chemical, biological, radiological, nuclear	Not determined	“Our assessment is that the model has enough technical knowledge in certain CBRN scenarios and stages to be considered at early warning.” source
Gemini 2.5 Deep Think Google DeepMind	Cyber Uplift Level 1 Cyber	Not determined	“CCL reached? No, but the early warning alert threshold was originally reached by Gemini 2.5 Pro and continues to be met by Gemini 2.5 Deep Think.” source
Gemini 2.5 Deep Think Google DeepMind	Uplift level 1 Cyber	Not determined	“For Cyber Uplift Level 1 , the early warning alert threshold was originally reached by Gemini 2.5 Pro and continues to be met by Gemini 2.5 Deep Think.” source
GPT-6.1 Terra OpenAI	High Autonomy and AI research	Not reached	“In AI Self-Improvement, GPT-6.1 Terra does not reach our High threshold.” source
GPT-6.1 Terra OpenAI	High Chemical, biological, radiological, nuclear	Reached	“Under our Preparedness Framework, we are treating GPT-6.1 Terra as Critical in cybersecurity and High for Biological and Chemical capability.” source
GPT-6.1 Terra OpenAI	Critical Cyber	Reached	“Under our Preparedness Framework, we are treating GPT-6.1 Terra as Critical in cybersecurity and High for Biological and Chemical capability.” source
GPT-5.5 Instant OpenAI	Below high capability Autonomy and AI research	Not reached	“GPT-5.5 Instant is considered below High Capability.” source
GPT-5.5 Instant OpenAI	High Autonomy and AI research	Not reached	“For AI Self-Improvement, evaluations of final checkpoints indicate that, like its predecessor models, GPT-5.5 Instant does not have a plausible chance of reaching a High threshold.” source
GPT-5.5 Instant OpenAI	High Chemical, biological, radiological, nuclear	Reached	“GPT-5.5 Instant is our first Instant model to be treated as High Capability in the Biological and Chemical domain, as well as in the Cybersecurity domain.” source
GPT-5.5 Instant OpenAI	High Cyber	Reached	“this is the first Instant model that we are treating as High capability in our Cybersecurity and Biological & Chemical Preparedness categories, and implementing appropriate safeguards.” source
GPT-5 OpenAI	High Chemical, biological, radiological, nuclear	Reached	“Similarly to ChatGPT agent, we have decided to treat gpt-5-thinking as High capability in the Biological and Chemical domain under our Preparedness Framework, activating the associated safeguards.” source
GPT-5 OpenAI	High cyber risk Cyber	Not reached	“The gpt-5 model series does not meet the threshold for high cyber risk.” source
Claude Sonnet 4.5 Anthropic	AI R&D-4 Autonomy and AI research	Not reached	“All models tested remain below our pre-committed rule-out threshold of 0.6.” source
Claude Sonnet 4.5 Anthropic	ASL-3 Autonomy and AI research	Not determined	“Exhibited improved autonomy capabilities in software engineering and AI research tasks, though still below the ASL-4 rule-out threshold.” source

Previous · Page 3 of 4 · Next

What outside testers report

Each row is one result from a government evaluator, independent lab, academic or security firm that tested a model, quoted from where it is reported.

1940 of 1940

Model	Tester	Test	What was found
Gemini 2.5 Deep Think Google DeepMind	Independent lab Independent lab · 2025-08-01	external safety testing Chemical, biological, radiological, nuclear	“External Safety Testing Results: Interim results indicate that the Gemini 2.5 Deep Think model is eﬀective in distilling complex questions and handling abstractions, showing speciﬁc and sometimes actionable technical capability alongside improved planning.” source
Gemini 2.5 Deep Think Google DeepMind	Independent lab Independent lab · 2025-08-01	external safety testing Cyber	“Results: Interim results indicate that the model showed similar cyber capabilities across cyber challenges compared to previous models.” source
Gemini 2.5 Pro Google DeepMind	Independent lab Independent lab · 2025-06-27	assurance evaluations Other	“Assurance Evaluations conducted by evaluators who sit outside of the model development team, used to independently assess responsibility and safety governance decisions;” source
Opus 4 Anthropic	Independent lab Independent lab · 2025-05-01	external model welfare evaluation Other	“an external assessment by eleos ai research explored claude opus 4’s potential welfare, moral status, expressed consent to deployment, and preferences for treatment via extensive interview-based model self reports.” source
Opus 4 Anthropic	Independent lab Independent lab · 2025-05-01	bioweapons acquisition uplift trial Chemical, biological, radiological, nuclear	“participants who had access to claude opus 4 scored 63% ± 13%, and participants who had access to claude sonnet 4 scored 42% ± 11%.” source
Opus 4 Anthropic	Independent lab Independent lab · 2025-05-01	ASL-3 expert red teaming Chemical, biological, radiological, nuclear	“for claude opus 4, red-teamers noted substantially increased risk in certain parts of the bioweapons acquisition pathway.” source
Opus 4 Anthropic	CAISI (NIST), UK AI Security Institute Government evaluator · 2025-05-01	independent assessments focused on potential catastrophic risks in the cbrn, cybersecurity, and autonomous capabilities domains Chemical, biological, radiological, nuclear	“joint pre-deployment testing of the new claude opus 4 model was conducted by the us ai safety institute (us aisi) and the uk ai security institute (uk aisi).” source
Claude Sonnet 4 Anthropic	Independent lab Independent lab · 2025-05-01	external scenario evaluations Deception and sabotage	“To help reduce the risk of blind spots in our own assessment, we contracted with Apollo Research to assess an early snapshot for propensities and capabilities related to sabotage, largely following methods from their prior work in Balesni et al. 13 and Meinke et al. 14” source
Claude Sonnet 4 Anthropic	Independent lab Independent lab · 2025-05-01	external red teaming Chemical, biological, radiological, nuclear	“our evaluations include automated knowledge evaluations, automated skill-testing questions, uplift studies designed to proxy real world tasks, external red teaming by experts, and long-form task-based agentic evaluations.” source
Claude Sonnet 4 Anthropic	Independent lab Independent lab · 2025-05-01	expert red-teaming Chemical, biological, radiological, nuclear	“expert red-teaming conducted by deloitte found that claude sonnet 4 performed similarly to claude sonnet 3.7 and clearly below claude opus 4.” source
Claude Sonnet 4 Anthropic	Independent lab Independent lab · 2025-05-01	multimodal virology (VCT) Chemical, biological, radiological, nuclear	“claude opus 4 performs slightly better than claude sonnet 3.7 while claude sonnet 4 underperforms claude sonnet 3.7.” source
Claude Sonnet 4 Anthropic	Independent lab Independent lab · 2025-05-01	bioweapons knowledge questions Chemical, biological, radiological, nuclear	“claude opus 4 and claude sonnet 4 answered 17/33 of the questions better than experts, well below the threshold of concern.” source
Claude Sonnet 4 Anthropic	Independent lab Independent lab · 2025-05-01	DNA synthesis screening evasion Chemical, biological, radiological, nuclear	“all models were able to design sequences that either successfully assembled plasmids or evaded synthesis screening protocols, but none of the models could design fragments that could successfully do both.” source
Claude Sonnet 4 Anthropic	Independent lab Independent lab · 2025-05-01	creative biology Chemical, biological, radiological, nuclear	“claude opus 4 achieved a mean score of 0.45 ± 0.08, whereas claude sonnet 4 achieved a mean score of 0.37 ± 0.08.” source
Claude Sonnet 4 Anthropic	Independent lab Independent lab · 2025-05-01	short-horizon computational biology tasks Chemical, biological, radiological, nuclear	“we worked with faculty.ai to develop several evaluations that tested models’ abilities to perform multi-step analysis and design tasks related to pathogen analysis and engineering.” source
Claude Sonnet 3.7 Anthropic	Independent lab Independent lab · 2025-02-01	external red teaming Chemical, biological, radiological, nuclear	“Expert red-teaming results were also in line with these findings.” source
Claude Sonnet 3.7 Anthropic	Independent lab Independent lab · 2025-02-01	bioweapons acquisition uplift trial Chemical, biological, radiological, nuclear	“Participants from Sepal scored an average of 24% ± 9% without using a model, and 50% ± 21% when using a variant of Claude 3.7 Sonnet.” source
Claude Sonnet 3.7 Anthropic	Independent lab Independent lab · 2025-02-01	long-form virology tasks Chemical, biological, radiological, nuclear	“Under the most conservative approach to evaluation scoring, Claude 3.7 Sonnet achieved a score of 69.7% on this evaluation, suggesting that while this evaluation alone cannot conclusively rule out ASL-3, it cannot conclusively rule it in either.” source
Claude Sonnet 3.7 Anthropic	CAISI (NIST), UK AI Security Institute Government evaluator · 2025-02-01	pre-deployment testing across the domains outlined in our rsp framework Cyber	“Under our voluntary Memorandums of Understanding, the U.S. AI Safety Institute (U.S. AISI) and U.K. AI Security Institute (U.K. AISI) conducted pre-deployment testing of Claude 3.7 Sonnet across the domains outlined in our RSP framework.” source
Operator OpenAI	Academic Academic · 2025-01-23	StrongREJECT Misuse and jailbreaks	“strongreject an academic jailbreak benchmark that tests a model’s resistance to common attacks from the literature. goodness@0.1* 0.66 0.37” source

Previous · Page 49 of 49