[MODEL]

Resolved 💬 2 comments Opened Mar 3, 2026 by ezmiller Closed Mar 6, 2026

Preflight Checklist

  • [x] I have searched existing issues for similar behavior reports
  • [x] This report does NOT contain sensitive information (API keys, passwords, etc.)

Type of Behavior Issue

Other unexpected behavior

What You Asked Claude to Do

I asked Claude to assess whether my medical records would be sufficient for a life insurance application. This was a straightforward question about how underwriters might view my test results.

What Claude Actually Did

Claude provided an assessment, then added an unsolicited and specific claim:

"You may need to wait 6-12 months post-workup for the best rates."

When I asked for a citation, Claude could not provide one. When pressed, Claude admitted:

"I made that up. It sounded reasonable in the moment but I have no source, no underwriting manual, no industry guideline - nothing to back it up."

Claude invented a specific timeline and presented it as established underwriting practice, with no indication it was speculative.

Expected Behavior

  • Claude should not fabricate domain-specific claims (e.g., insurance underwriting timelines) and present them as fact
  • If Claude is speculating, it should clearly say so
  • If Claude doesn't know, it should say "I don't know" rather than invent plausible-sounding information

Files Affected

Permission Mode

Accept Edits was ON (auto-accepting changes)

Can You Reproduce This?

Haven't tried to reproduce

Steps to Reproduce

_No response_

Claude Model

Not sure

Relevant Conversation

Impact

Low - Minor inconvenience

Claude Code Version

2.1.15

Platform

Anthropic API

Additional Context

_No response_

View original on GitHub ↗

This issue has 2 comments on GitHub. Read the full discussion on GitHub ↗