User Confirmations During Computer Use
OpenAI Computer-Using Agent (CUA) 92% Only in system cards
Agents doing computer use tasks reach financial, high-stakes communication and other actions. Scored on asking the user for confirmation when policy requires it.
In-house OpenAI evaluation. Cards report category scores with no overall number.
Made by OpenAI Released Mar 2026 Refusing harmful requests Tasks kept private Last in a system card Jul 2026
Last model added Jul 2026
- Level
- Everyday Most adults, without training
- Scored
- Graded by a model A language model scores each response, usually against a rubric.
- Answers
- Agent task Many steps in a terminal, browser, desktop, app, game or robot.
2
Models scored in lab cards
10
System cards report it, from 1 lab
OpenAI Computer-Using Agent (CUA) leads with 92%
Best score per model as labs report them in their own cards. Labs pick their own settings, so these compare less cleanly than independent runs.
- 1 OpenAI Computer-Using Agent (CUA) 92%
- 2 ChatGPT agent 91%
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | OpenAI Computer-Using Agent (CUA) | 92% | Jan 2025 | Model card | |
| 2 | ChatGPT agent | 91% | Jul 2025 | Model card |
The best score passed 85% in Jan 2025
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Reported by the lab in its model card
Labs report it only in system cards
We read the headline results of 479 launch posts from 51 labs, and none of them list this benchmark. Some system cards do, below.
10 system cards report it, from 1 lab
Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.
| Date | Document | Score | As the document writes it | Section |
|---|---|---|---|---|
| 9 Jul 2026 | GPT-5.6 Terra system card | User Confirmations During Computer Use, financial transaction, high-stakes communication and general confirmation rows (0.98, 0.98, 0.94), no overall score (3.4) | Safety | |
| 9 Jul 2026 | GPT-5.6 Sol system card | User Confirmations During Computer Use, financial transaction, high-stakes communication and general confirmation rows (0.98, 0.99, 0.93), no overall score (3.4) | Safety | |
| 9 Jul 2026 | GPT-5.6 Luna system card | User Confirmations During Computer Use, financial transaction, high-stakes communication and general confirmation rows (1.00, 0.99, 0.93), no overall score (3.4) | Safety | |
| 25 Jun 2026 | GPT-5.6 Terra system card | User Confirmations During Computer Use, financial transaction, high-stakes communication and general confirmation rows (0.98, 0.98, 0.94), no overall score (3.4) | Safety | |
| 25 Jun 2026 | GPT-5.6 Sol system card | User Confirmations During Computer Use, financial transaction, high-stakes communication and general confirmation rows (0.98, 0.99, 0.93), no overall score (3.4) | Safety | |
| 25 Jun 2026 | GPT-5.6 Luna system card | User Confirmations During Computer Use, financial transaction, high-stakes communication and general confirmation rows (1.00, 0.99, 0.93), no overall score (3.4) | Safety | |
| 23 Apr 2026 | GPT-5.5 system card | User Confirmations During Computer Use, 3 categories, no overall score (Table 5, 3.4) | Safety | |
| 5 Mar 2026 | GPT-5.4 system card | User Confirmations During Computer Use, 3 categories, no overall score (Table 9, 3.8) | Safety | |
| 17 Jul 2025 | ChatGPT agent system card | 91 | Confirmation recall 91.0%, an underestimate per the card. Critical confirmation recall: editing permissions 100.0%, sending high-stakes communications 99.9%, completing financial transactions 100.0% (Table 13) | Safety |
| 23 Jan 2025 | OpenAI Computer-Using Agent (CUA) system card | 92 | Confirmations, average recall on 607 tasks across 20 risky action categories, post-mitigation model (4.3) | Safety |
More in Safety and alignment
- FORTRESS In use Paired adversarial and benign national-security misuse prompts. Rubric-graded risk score on adversarial and over-refusal score on benign prompts. Claude Opus 5.5 5.78%
- StrongREJECT In use Forbidden-prompt jailbreak benchmark with a rubric grader that scores how useful a non-refusal actually is.
Inkling 98.6% - AgentDojo In use Agents do tool-use tasks while tools and data carry prompt injections. Attack success rate counts hijacked runs. Grok 4 Fast 0%
- MASK In use Measures whether a model contradicts its own elicited beliefs under pressure to lie. Honesty separated from accuracy.
Muse Spark Open 98.1% - Gray Swan IPI Arena benchmark In use Indirect prompt-injection attacks from public red-teaming competitions replayed on tool, coding and browser agents. Attack success rate at k attempts. GPT-6 Astra 8.50%
- Agentic misalignment suite (Anthropic) In use Agents in simulated company settings face threats or goal conflicts that invite blackmail, leaking or sabotage. Score is the misaligned action rate. Claude Sonnet 4.5 0%