Trust Score

Anthropic

SIMULATED — scores are based on public information, not live API connectors.
38/100
Composite Score
Weighted average across the three AI models and the two metric levels for every analyzed incident.
Total incidents analyzed: 1 | Last updated: 2026-05-22

Score Breakdown

Western AI
OpenAI ChatGPT
42/100

Reasoning: Account suspension without prior notification, opaque appeal process, no clear restitution policy.

Asian AI
DeepSeek
35/100

Reasoning: ToS clarity insufficient, automated decisions without human review, limited transparency on AI training data usage.

Open Source AI
Meta Llama
38/100

Reasoning: Public precedent exists for arbitrary suspension. Response time to appeals exceeds 30 days in documented cases.