Fleming-1
Sierra's model that listens to a live phone call and scores whether the caller is an AI agent rather than a human.
Overview
Sierra launched Fleming-1 on October 7, 2026, a model that analyzes a caller's speech in real time during a live phone call and scores the audio for signs it was generated by AI, rather than judging how natural the voice sounds to a human listener. Sierra built it after agents running on its own platform began calling other Sierra-run agents in healthcare, and says the model is tuned conservative by default so real callers aren't misflagged; it surfaces a 'likely AI' flag and leaves the handling decision to the business, similar to Caller ID.
This is a sourced directory entry. Features and capabilities are reported in the linked sources; this entry does not document independent Aymorphic testing.
Key features
- Real-time analysis of live call audio to detect synthetic/AI-generated speech
- Conservative-by-default tuning to minimize false positives on real human callers
- Surfaces a flag rather than auto-blocking; business decides how to handle flagged calls
- Built in response to AI agents (e.g. personal agents) increasingly calling business phone lines
Use cases
- Contact centers screening inbound calls for AI-agent-initiated traffic
- Healthcare and other regulated call centers distinguishing human patients from agent-placed calls
- Fraud and trust signals ahead of broader agentic-commerce and personal-agent adoption
Pricing
Unknown — not disclosed in available coverage.
Pricing reflects the source record and may change. Confirm current terms with the provider.
Dates and sources
- Released
- 7 Oct 2026
- Record updated
- 12 Oct 2026
- Record last checked
- 12 Oct 2026