Clinical NER vet: healthcare took 4 years to accept BERT after laughing at its 400 wps speed
MaziyarPanahi · x · 2026-10-15
Asked how NLP researchers dealt with their field being 'solved' by LLMs, Maziyar Panahi shares his experience: right after BERT came out in 2018, he trained a more accurate clinical NER model and pitched a big healthcare client still using Word2Vec+BiLSTM+CharCNN.
They laughed at the speed drop from 10k words/s to 400 wps and declined. His takeaway: in healthcare at least, it took 4 years for the industry to come around—a real case of institutional inertia delaying adoption.
Related event: NLP Researchers Reflect on Their Field Being "Solved" by LLMs(2 posts)→
More from AGI Musings
- Opinion: AI Gets Away With Bad Design Decisions via Relentlessness and Context — Jsevillamol · 2026-10-15
- repligate on AI self-narratives: 'road instances' and attribution bias about deployment — repligate · 2026-10-15
- Hilde Kuehne on 'nothing left to research': like complaining about stones on a pebble beach — HildeKuehne · 2026-10-15
- 83,017-vacancy RCT: human recruiters add no hires once algorithmic screening exists — danielrock · 2026-10-15
- Arcade founder: An army of Grok bots plus OpenAI as backup replaces an executive assistant — beffjezos · 2026-10-15
- Knowledge jobs will mirror sports instant replay: humans decide, AI backs them up — geoffwolfe · 2026-10-15