← All verdicts

Can you trust Claude Haiku 5.5 for space work?

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

On its own, trust it for baseline knowledge. Connected to Orbit Sentinel, trust it for catalog lookup, debris & conjunction context, regulatory & licensing, company & market intelligence, baseline knowledge.

Job Alone With data
right / wrong declined right / wrong declined
Catalog lookup
27 / 24
49
100 / 0
0
Constellation & fleet status
26 / 16
58
87 / 12
2
Debris & conjunction context
19 / 10
71
93 / 0
7
Regulatory & licensing
23 / 8
69
82 / 3
16
Company & market intelligence
27 / 10
63
94 / 2
3
Ground infrastructure
13 / 3
85
89 / 6
5
Baseline knowledge
100 / 0
0
97 / 0
3
Orbit Sentinel-only questions
0 / 0
100
100 / 0
0

% 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.

How it was run
Claude Code on a Claude Pro subscription
Tool loop
Claude Code's own tool loop
Runs
1 on its own, 3 with tools; rates pool all runs, ranges count each question once
Verdict rule
Trust it: wrong ≤ 5% and right ≥ 80%. Don't use it: wrong > 20%. Otherwise verify its answers.

Limits, harness differences and corrections: methodology.