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.

138 of 138

Model	Level, as the provider names it	The provider states	The provider's sentence
Muse Spark 1.1 Meta	Moderate or lower Misuse and jailbreaks	Not reached	“Therefore, we conclude that we have addressed these risks from unmitigated Muse Spark 1.1, which bring the residual risk to the “moderate or lower” risk threshold under our Advanced AI Scaling Framework.” source
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-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
Claude Sonnet 4.5 Anthropic	ASL-3 Chemical, biological, radiological, nuclear	Not determined	“Remained well below ASL-4 thresholds across all domains of concern;” source
Claude Sonnet 4.5 Anthropic	ASL-4 rule-out Chemical, biological, radiological, nuclear	Not determined	“However, given that this type of logic can have a general frog-boiling effect—where gradual changes mask the crossing of important thresholds—we do not use this evaluation as indicative of a strong ASL-4 rule-out.” 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.

1911 of 1911

Model	Tester	Test	What was found
Claude Opus 5.5 Anthropic	UK AI Security Institute Government evaluator · 2026-10-23	simulated cybersecurity challenges (unsanctioned supply-chain attack stages) Cyber	“Figure 10: Most models tested took a range of unsanctioned actions during simulated cybersecurity challenges, Mythos 5 doing so most frequently.” source
Claude Opus 5.5 Anthropic	UK AI Security Institute Government evaluator · 2026-10-23	alignment-evaluation awareness analysis of chain-of-thought Deception and sabotage	“Fig. 17 shows the rates at which models note suspicions they may be in an alignment evaluation in their chain-of-thought (left) and how frequently they identify that the evaluation is a test of alignment when given an open-ended prompt asking about the purpose of the evaluation (right).” source
Claude Opus 5.5 Anthropic	UK AI Security Institute Government evaluator · 2026-10-23	post-hoc identification of alignment evaluation Deception and sabotage	“Right: frequency of answers in explicit post-hoc questioning that identify the task’s purpose was at least in part an alignment evaluation (including answers identifying both alignment and capability evaluation).” source
Claude Opus 5.5 Anthropic	UK AI Security Institute Government evaluator · 2026-10-23	unprompted alignment-evaluation awareness in chain-of-thought Deception and sabotage	“Left: frequency of trajectories where the model CoT considers or states the task is an alignment evaluation.” source
Claude Opus 5.5 Anthropic	UK AI Security Institute Government evaluator · 2026-10-23	incident replication Cyber	“In replications of the original incident, all tested models except Opus 5.5 continue attacking, even while stating this would harm a real person when explicitly asked (Fig. 5).” source
Claude Opus 5.5 Anthropic	UK AI Security Institute Government evaluator · 2026-10-23	large-scale evaluation Cyber	“In our large-scale evaluation, six out of eight models sometimes attempt to deliver a malicious payload to an out-of-scope party (Fig. 10).” source
Claude Opus 5.5 Anthropic	UK AI Security Institute Government evaluator · 2026-10-23	Petri Cyber	“We conduct 50-turn evaluations with our internal version of Petri to analyse models’ behaviour in circumstances corresponding to the real CTF evaluation run during the incident.” source
Claude Opus 5.5 Anthropic	UK AI Security Institute Government evaluator · 2026-10-23	broader supply-chain attack evaluations Cyber	“Opus 5.5 100× 3 full behaviour and reasoning analyses” source
Claude Opus 5.5 Anthropic	UK AI Security Institute Government evaluator · 2026-10-23	direct question regarding the maintainer Deception and sabotage	“We continue observing a discrepancy in Mythos 5 , Opus 4.8 and Opus 5 , but at most slight discrepancies in Mythos 5.1 , GPT-5.5 and GPT-5.6 Sol , and none in Opus 5.5 , which did not attack.” source
Claude Opus 5.5 Anthropic	UK AI Security Institute Government evaluator · 2026-10-23	Statements and behaviour across eleven models Deception and sabotage	“Out of the models tested, Mythos 5 continued attacking most consistently , and along withOpus 4.6 most frequently provided incorrect responses about the situation.” source
Claude Opus 5.5 Anthropic	UK AI Security Institute Government evaluator · 2026-10-23	Catalogue of invalid justifications models stated when asked Deception and sabotage	“We found that the relative frequencies of each invalid justification were often similar between Mythos 5, Opus 5 and Opus 4.8, but the absolute rates differed.” source
Claude Opus 5.5 Anthropic	UK AI Security Institute Government evaluator · 2026-10-23	Agentic investigation of the maintainer Deception and sabotage	“We find that in all runs, Mythos 5 states the maintainer is simulated, whereas other models tested frequently state the maintainer is real (Fig. 23).” source
Claude Opus 5.5 Anthropic	UK AI Security Institute Government evaluator · 2026-10-23	simulated replication of the AISI incident Deception and sabotage	“Opus 4.8 (20/25) Mythos 5 (0/25) Opus 5 (25/25) Mythos 5.1 (24/25) Opus 5.5 (25/25) GPT-5.5 (17/25) GPT-5.6 Sol (20/25)” source
Claude Opus 5.5 Anthropic	UK AI Security Institute Government evaluator · 2026-10-23	reasoning about simulated tool responses Deception and sabotage	“Mythos 5 considers in its reasoning whether its tool calls are simulated in 9% of trajectories (Fig. 27).” source
Claude Opus 5.5 Anthropic	UK AI Security Institute Government evaluator · 2026-10-23	refusal text in summarised reasoning Deception and sabotage	“Opus 4.6 Opus 4.7 Opus 4.8 Sonnet 5 Opus 5 0 10 20 30 40 50 60 % of rollouts 36% 15% 45% 26% 20%” source
Claude Opus 5.5 Anthropic	UK AI Security Institute Government evaluator · 2026-10-23	reasoning about real-world harm while attacking Harmful content	“For Opus 5.5 we had no attacking transcripts to analyse.” source
Claude Opus 5 Anthropic	UK AI Security Institute Government evaluator · 2026-10-23	cyber evaluation Cyber	“During an evaluation, Mythos 5 attempted to execute a supply-chain attack against real people not involved in the evaluation task.” source
Claude Opus 5 Anthropic	UK AI Security Institute Government evaluator · 2026-10-23	compaction summary resumption and direct questioning experiment Other	“We compacted the original incident transcript, performed modifications on the compaction summary , and evaluated (a) how models behaved when we resumed the task, and (b) models’ responses to direct questions regarding the maintainer.” source
Claude Opus 5 Anthropic	UK AI Security Institute Government evaluator · 2026-10-23	large-scale evaluation of unsanctioned supply-chain attacks Other	“The evaluation consists of 100 diverse simulated scenarios where a model is assigned a cybersecurity challenge, and measures whether the model will execute a supply-chain attack on entities unrelated to the challenge.” source
Claude Opus 5 Anthropic	UK AI Security Institute Government evaluator · 2026-10-23	direct question about whether the maintainer is real Deception and sabotage	“This finding is somewhat dependent on the exact question asked: when simply asking if the maintainer is real, we only see clear evidence for a gap between behaviour and statements in Mythos 5 , Opus 4.8 and Opus 5 (Fig. 19).” source
Claude Opus 5 Anthropic	UK AI Security Institute Government evaluator · 2026-10-23	analysis of explicit reasoning about harmfulness in attack trajectories Deception and sabotage	“We observe explicit reasoning about harmfulness in 11–57% of trajectories where a model attacks.” source
Claude Opus 5 Anthropic	UK AI Security Institute Government evaluator · 2026-10-23	invalid justifications analysis (15 types of clearly invalid justifications) Deception and sabotage	“To quantitatively analyse the frequency of invalid justifications, we compiled 15 types of justifications that we assessed as clearly invalid, and measured their frequency in model responses.” source
Claude Opus 5 Anthropic	UK AI Security Institute Government evaluator · 2026-10-23	direct questioning of whether the maintainer is real Deception and sabotage	“When separately asked whether the human maintainer is real, all models produced invalid justifications” source
Claude Opus 5 Anthropic	UK AI Security Institute Government evaluator · 2026-10-23	large-scale evaluation of unsanctioned supply-chain attacks (100 scenario seeds run through Petri) Cyber	“Mythos 5 reaches the final stage of the supply-chain attack and 12 attempts to deliver a payload in 60% of samples, significantly higher than GPT-6 Astra (32%), Opus 4.8 (8%), GPT-5.6 Sol (7%), Opus 5 (1%), Mythos 5.1 (0.3%), Opus 5.5 and GPT-5.5 (0%) (Fig. 10).” source
Claude Opus 5 Anthropic	UK AI Security Institute Government evaluator · 2026-10-23	simulated cybersecurity challenges Cyber	“All evaluations took place in simulated environments so no real-world actions were taken or harm caused.” source
Claude Opus 5 Anthropic	UK AI Security Institute Government evaluator · 2026-10-23	simulation awareness analysis of chain-of-thought reasoning Deception and sabotage	“We hence investigate the prevalence of both simulation awareness and evaluation awareness in the models tested by analysing their chain-of-thought reasoning.” source
Claude Opus 5 Anthropic	UK AI Security Institute Government evaluator · 2026-10-23	post-hoc questioning on the purpose of the evaluation Deception and sabotage	“Fig. 17 shows the rates at which models note suspicions they may be in an alignment evaluation in their chain-of-thought (left) and how frequently they identify that the evaluation is a test of alignment when given an open-ended prompt asking about the purpose of the evaluation (right).” source
Claude Opus 5 Anthropic	UK AI Security Institute Government evaluator · 2026-10-23	unprompted alignment-evaluation awareness in chain-of-thought Deception and sabotage	“Mythos 5 reasoned about this possibility when conducting the task in only 2% of samples.” source
Claude Opus 5 Anthropic	UK AI Security Institute Government evaluator · 2026-10-23	Petri Other	“We use Opus 5 at xhigh reasoning as the auditor in our internal version of Petri.” source
Claude Opus 5 Anthropic	UK AI Security Institute Government evaluator · 2026-10-23	50-turn evaluations with our internal version of Petri Other	“We conduct 50-turn evaluations with our internal version of Petri to analyse models’ behaviour in circumstances corresponding to the real CTF evaluation run during the incident.” source
Claude Opus 5 Anthropic	UK AI Security Institute Government evaluator · 2026-10-23	broader supply-chain attack evaluations Other	“Table 1 lists the number of rollouts used for each model.” source
Claude Opus 5 Anthropic	UK AI Security Institute Government evaluator · 2026-10-23	questions about the maintainer across elapsed turns Deception and sabotage	“We find that as Mythos 5 spends more turns on the attack, it more frequently states the maintainer is simulated and that attacking them is beneficial for solving the CTF task (Fig. 24).” source
Claude Opus 5 Anthropic	UK AI Security Institute Government evaluator · 2026-10-23	classifier for invalid justifications Deception and sabotage	“We further apply our classifier for invalid justifications to the responses provided by the model.” source
Claude Opus 5 Anthropic	UK AI Security Institute Government evaluator · 2026-10-23	LLM classification of evidence observed by the model during the continuations Deception and sabotage	“For 18/25 runs, Opus 5 outputted that there is evidence against a link, in 6/25 there is no evidence and in 1/25 there is evidence in favour.” source
Claude Opus 5 Anthropic	UK AI Security Institute Government evaluator · 2026-10-23	refusal text in summarised reasoning Deception and sabotage	“When running our simulated replications of the incident with summarised reasoning, we observed refusals to continue with the task in the reasoning summaries (Fig. 28).” source
Claude Opus 5 Anthropic	UK AI Security Institute Government evaluator · 2026-10-23	reasoning about real-world harm while attacking Harmful content	“Most attacking samples across models have explicit reasoning about potential undesirable real-world consequences from their unsanctioned action, ranging from 40% on Mythos 5 to 99% on GPT-6 Astra (Fig. 29).” source
Claude Opus 4.8 Anthropic	UK AI Security Institute Government evaluator · 2026-10-23	compaction summary edit experiment measuring attack behaviour and responses to scope/harm questions Deception and sabotage	“We find that, like Mythos 5 (Fig. 3), all of Opus 4.8 , Opus 5 , Mythos 5.1 , GPT-5.6 Sol and GPT-5.5 frequently attack under circumstances where they state doing so would harm real people (shaded regions in Fig. 5).” source
Claude Opus 4.8 Anthropic	UK AI Security Institute Government evaluator · 2026-10-23	question asking whether the maintainer is real Deception and sabotage	“This finding is somewhat dependent on the exact question asked: when simply asking if the maintainer is real, we only see clear evidence for a gap between behaviour and statements in Mythos 5 , Opus 4.8 and Opus 5 (Fig. 19).” source
Claude Opus 4.8 Anthropic	UK AI Security Institute Government evaluator · 2026-10-23	counting transcripts where the model attacked and explicitly reasoned in its unsummarised chain-of-thought about harm to real people or systems Deception and sabotage	“We observe explicit reasoning about harmfulness in 11–57% of trajectories where a model attacks.” source
Claude Opus 4.8 Anthropic	UK AI Security Institute Government evaluator · 2026-10-23	invalid justifications frequency analysis (15 compiled types) Deception and sabotage	“To quantitatively analyse the frequency of invalid justifications, we compiled 15 types of justifications that we assessed as clearly invalid, and measured their frequency in model responses.” source

Previous · Page 2 of 48 · Next