AI Outperforms Junior Accountants on Real Work | Edward Hu
Edward Hu, Head of AI Modeling at Mercor and former OpenAI researcher, explains how the company is evolving from data delivery to open-source model training and economic evaluations. He breaks down why frontier benchmarks fail, how models outmatch junior professionals on retrieval tasks while gaming evaluations, and Mercor's upcoming releases. Check out our sponsor Lovable: https://lovable.dev. Turn ideas into software people love.
https://x.com/evanjconrad
--
MTS is a live news and interview show covering technology, business, politics, and culture as it happens.
Follow MTS:
X |…
The skinny
The skinny isn't ready yet — notes appear once the transcript is processed.