Sounds Right, Isn’t: Where Frontier AI Models Fall Short on Employee Feedback
Companies are handing frontier AI some of their most sensitive work — reading employee feedback and turning it into decisions about people — with no real way to know whether the models are any good at it. It’s part of a bigger shift: AI is fast becoming part of how organizations understand and act on the experience of their people, and the stakes keep rising.
In the first PYX Labs webcast, Joe Freed (Founder, PYX Labs) and Sarah Foster (Director, PYX Labs) host a conversation that goes wider than the benchmark itself, joined by Melissa Valentine, PhD, of Stanford and the Stanford Institute for Human-Centered AI (HAI). They’ll take on why getting AI right for the workplace matters so much, what Melissa’s lab is learning about how AI is reshaping work and organizations, and the real risks and opportunities of using it in employee listening. At the center is PYX-Voice, PYX Labs’ first benchmark: how seven frontier models handled 84 real employee-experience tasks, where they fell short of the expert bar, and what that reveals about where AI can be trusted with people. The conversation closes on what comes next, for the research and for the field.
- August 5, 2026
- 9:00 a.m. PT / 12:00 p.m. ET
- Live, roughly 45 minutes with audience Q&A
- Free to attend · Register to get the recording
What we'll cover
- Why this matters. The stakes of AI in the workplace, and the risks and opportunities of using it in employee listening. Melissa shares what her Stanford lab is learning about how AI is reshaping work, teams, and organizations.
- What we found. Sarah walks through PYX-Voice: how seven frontier models did on real employee-listening work, where they fell short of the expert bar, and why the gaps land where the stakes are highest — with Joe drawing out the takeaways along the way.
- Where this goes. What we learned, what Melissa is researching next, and what PYX Labs is building toward.
Hosts
Joe Freed
Joe leads the team building PYX Labs and the products enterprises rely on to understand their people.
Sarah Foster
An I-O psychologist, Sarah led the design of the PYX-Voice benchmark and the expert criteria it grades against.
Guest
Melissa Valentine, PhD
An academic advisor to PYX Labs, Melissa studies how AI systems reshape work, teams, and organizations.