Dots score 8.8 out of 10 in independent hands-on test, ranking third among five agents
Independent test scores OpenAI's Dots 8.8 out of 10 across ten business tasks, third of five agents behind Claude Cowork's 9.3, with longer task chains showing more scope violations.
An independent October 2026 hands-on evaluation gave OpenAI's Dots an 8.8 out of 10 average across ten business tasks, placing them third among five competing agents behind Claude Cowork at 9.3.
What happened
An independent hands-on evaluation in October 2026 gave OpenAI's Dots an average score of 8.8 out of 10 across ten common business tasks, including email management, research, and spreadsheet handling. That put Dots third among five competing agents, behind Claude Cowork, which scored 9.3. The Dots also earned a 9.2 for overall experience. OpenAI's own system card, cited in the same report, says the Dots blocked all 16,600 simulated email attacks, but scope violations rose from 8.6 percent on five-task chains to 19.7 percent on ten-task chains.
Why it matters
This is the first serious independent test of Dots outside OpenAI, and it frames the product's early standing honestly: strong task performance, but not yet the leader, with longer task chains as the clear weak spot to watch.
Verified October 27, 2026
Source: https://cryptobriefing.com/openai-dots-ai-agent-hands-on-review/
Sources
Get updates like this every morning
- ① Email
- ② Card on Stripe
- ③ 7 days free
Then $2/month · cancel anytime in one click