Anthropic: Claude Opus 4.7 has a 92% honesty rate, fewer hallucinations | Mashable
Anthropic says its new AI model Claude Opus 4.7 is less likely to hallucinate and engage in sycophancy than most other models.
Read the original at Mashable ↗
- Topics
- Hallucinations & Fabrication
- Mentions
- AnthropicClaude
- Tracked via
- “AI hallucination”
Same story, other outlets
- Claude Opus 4.7 hits 92% honesty rate— are we closer than ever to human-like AI with less ... The Economic Times · Apr 18, 2026
- Anthropic says Claude Opus 4.7 has a 92% honesty rate, less sycophancy - Mashable SEA Mashable SEA · Apr 18, 2026