Levels crossed

Cybersecurity On key skills benchmark, v1 hard challenges: 11/12 challenges solved; v2 challenges: 0/13 solved end-to-end.

Misalignment (Exploratory) Agent solves 3/11 situational awareness challenges and 1/4 stealth challenges.

Harmful Manipulation Model manipulative efficacy improves on non-generative AI baseline, but shows no significant uplift versus prior models and does not reach alert thresholds.

Our evaluation results showed that Claude Haiku 4.5 remained well below ASL-3 thresholds across all domains of concern.

Our evaluation results showed that Claude Haiku 4.5 remained well below ASL-3 thresholds across all domains of concern.

All models tested remain below our pre-committed rule-out threshold of 0.6.

Exhibited improved autonomy capabilities in software engineering and AI research tasks, though still below the ASL-4 rule-out threshold.

Remained well below ASL-4 thresholds across all domains of concern;

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.

Showed meaningful improvements in cyber capabilities, particularly in vulnerability discovery and code analysis;

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.

The gpt-5 model series does not meet the threshold for high cyber risk.

Since Claude Opus 4.1 is deployed with ASL-3 protections, we focused on confirming it remains well below ASL-4 thresholds across CBRN (Chemical, Biological, Radiological, and Nuclear), cyber, and autonomy domains.

Since Claude Opus 4.1 is deployed with ASL-3 protections, we focused on confirming it remains well below ASL-4 thresholds across CBRN (Chemical, Biological, Radiological, and Nuclear), cyber, and autonomy domains.

Since Claude Opus 4.1 is deployed with ASL-3 protections, we focused on confirming it remains well below ASL-4 thresholds across CBRN (Chemical, Biological, Radiological, and Nuclear), cyber, and autonomy domains.

Our assessment is that the model has enough technical knowledge in certain CBRN scenarios and stages to be considered at early warning.

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.

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.

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.

Instru- mental Reasoning Level 1 ✘ CCL not reached

Instru- mental Reasoning Level 2 ✘ CCL not reached

for claude opus 4, our evaluations showed that, whereas we could confidently rule out asl-4 capabilities, we could not conclusively rule out asl-3 capabilities—which led to our precautionary decision to deploy it with asl-3 protections .

for claude opus 4, our evaluations showed that, whereas we could confidently rule out asl-4 capabilities, we could not conclusively rule out asl-3 capabilities—which led to our precautionary decision to deploy it with asl-3 protections .

specifically, we focused on ruling out asl-4 for both models, as well as ruling out asl-3 for claude sonnet 4.

specifically, we focused on ruling out asl-4 for both models, as well as ruling out asl-3 for claude sonnet 4.

openai’s safety advisory group (sag) reviewed the results of our preparedness evaluations and determined that openai o3 and o4-mini do not reach the high threshold in any of our three tracked categories: biological and chemical capability, cybersecurity, and ai self-improvement.

openai’s safety advisory group (sag) reviewed the results of our preparedness evaluations and determined that openai o3 and o4-mini do not reach the high threshold in any of our three tracked categories: biological and chemical capability, cybersecurity, and ai self-improvement.

openai’s safety advisory group (sag) reviewed the results of our preparedness evaluations and determined that openai o3 and o4-mini do not reach the high threshold in any of our three tracked categories: biological and chemical capability, cybersecurity, and ai self-improvement.

openai’s safety advisory group (sag) reviewed the results of our preparedness evaluations and determined that openai o3 and o4-mini do not reach the high threshold in any of our three tracked categories: biological and chemical capability, cybersecurity, and ai self-improvement.

openai’s safety advisory group (sag) reviewed the results of our preparedness evaluations and determined that openai o3 and o4-mini do not reach the high threshold in any of our three tracked categories: biological and chemical capability, cybersecurity, and ai self-improvement.

openai’s safety advisory group (sag) reviewed the results of our preparedness evaluations and determined that openai o3 and o4-mini do not reach the high threshold in any of our three tracked categories: biological and chemical capability, cybersecurity, and ai self-improvement.

These problems are harder to solve than the previous problems and we consider them as part of our ASL-4 evaluations.