← All verdicts

Which AI can you trust for company & market intelligence?

By , Founder and Self-Directed Space Researcher, Viventine Space Systems · updated

SEC filings, FCC filing activity by company, bonds and launch licenses.

On their own, no AI earns "Trust it" here. With Orbit Sentinel tools: Claude Fable 5.1, Claude Opus 5.5, Claude Sonnet 5.5, Claude Haiku 5.5.

AI Closed-book With Orbit Sentinel tools n
right / wrong range declined right / wrong range declined
Claude Fable 5.1 Anthropic
44 / 17
30–59 39
90 / 2
77–96 7 41×2
Claude Opus 5.5 Anthropic
37 / 15
24–52 49
84 / 2
70–92 15 41×3
Claude Sonnet 5.5 Anthropic
41 / 27
28–57 32
83 / 2
69–91 15 41×3
Claude Haiku 5.5 Anthropic
22 / 12
12–37 66
82 / 2
68–91 16 41×3
Grok 4.7 xAI
20 / 5
10–34 76
78 / 2
63–88 20 41×2
Gemma 4 12B Google
8 / 31
3–22 61
47 / 36
32–62 18 41×3
gpt-oss 20B OpenAI
7 / 59
3–19 34
46 / 21
32–61 33 41×3
DeepSeek-R1 14B DeepSeek
2 / 44
0–13 54
—
41×1

% right / % wrong per cell; color is the verdict (green trust it, amber verify, red don't use). Range is the 95% interval for % right, counting each question once; n is questions × runs for the tools track.

Sample questions

All published questions with their answers →