Tag
Round 3 of a hands-on comparison of 7 AI phone agents tests interruption handling and phone menu navigation, with Retell standing out for natural voice among build-it-yourself platforms and PlaceCall and ClawCall leading the 'just call' group.
The article reports on real-world testing of OpenAI's GPT-Live-1 model for phone agents, highlighting issues with instruction adherence, alphanumeric errors, and language handling in production scenarios.
PhoneHarness is a mixed-action benchmark and execution framework that evaluates phone-use agents on verifiable mobile workflows, achieving a 75% pass rate and outperforming existing approaches by 12.9 percentage points through deterministic action routing and auditable execution traces.