Call Center Jobs Are Declining, but 14,000 Calls Show Humans Still Outperform AI

Bold Agent Kit compared ~14,000 inbound calls (6,343 AI-handled, ~7,657 human-handled) to measure whether AI voice agents actually outperform human receptionists. The answer isn't a clean yes.

Key findings

• Human receptionists delivered ~4x stronger inbound intake conversion. Humans were far better at converting inquiries into successful intake outcomes — understanding needs, building trust, collecting accurate info, and moving callers to the next step.
• AI-handled calls had 700% more no-progress hangups. Callers hung up before meaningful progress was made at roughly 7x the rate seen on human-handled calls. The study didn't isolate the cause — conversational friction, response delays, unexpected questions, and caller preference for humans are all plausible.
• Human receptionists missed ~15% more calls. AI's real advantage is availability: it can answer consistently when staff aren't available. But answering more calls isn't the same as producing more business — an automated system that answers everything but can't convert may not deliver the expected ROI.

Why this matters for builders

If you're wiring up AI receptionists or voice agents, the study suggests benchmarking on outcomes (intake conversion, abandoned-without-progress rate), not just call-answer rate. A system that always picks up but leaks callers mid-intake will look good on uptime dashboards and bad on revenue.

The a16z October 9, 2026 report cited Revelio employment data showing global call-center employment growth turning negative across income groups — the first sustained decline after more than a decade of expansion. Whether AI caused it or not, the labor shift is measurable. What this study adds is that performance parity isn't a given, at least in the workflows Bold Agent Kit examined.

Full methodology and the 7-minute writeup are at the source link below.