← All verdicts

Which AI can you trust for ground infrastructure?

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

Earth stations, gateways, call signs and where they are.

On their own, no AI earns "Trust it" here. With Orbit Sentinel tools, none earns "Trust it" yet.

AI Closed-book With Orbit Sentinel tools n
right / wrong range declined right / wrong range declined
Claude Sonnet 5.5 Anthropic
18 / 27
7–39 55
74 / 17
53–88 9 22×3
Claude Fable 5.1 Anthropic
27 / 23
13–48 50
73 / 14
52–87 14 22×2
Claude Opus 5.5 Anthropic
18 / 23
7–39 59
68 / 5
47–84 27 22×3
Grok 4.7 xAI
14 / 5
5–33 82
57 / 11
37–75 32 22×2
Claude Haiku 5.5 Anthropic
5 / 9
1–22 86
53 / 14
33–72 33 22×3
gpt-oss 20B OpenAI
9 / 45
3–28 45
23 / 18
10–43 59 22×3
Gemma 4 12B Google
0 / 45
0–15 55
19 / 58
8–39 23 22×3
DeepSeek-R1 14B DeepSeek
14 / 36
5–33 50
—
22×1

% 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 →