Omar Khattab on impactful AI research: invest in projects, not papers

lateinteraction · x · 2026-11-17

lateinteraction cites Omar Khattab's blog post "On Impactful AI Research," endorsing Dimitris Papailiopoulos's point that most people agree the paper-counting machine is broken but few say it publicly because awards, promotions, and hiring depend on it. Khattab's advice for grad students: invest in projects, not papers; pick timely problems with headroom and fanout; think two steps ahead and iterate fast; ship your work and own popularizing it; funnel the excitement into growing open-source research; keep investing via new papers. Real impact comes from open-source artifacts—models, systems, frameworks, benchmarks—not论文 produced for the counting mechanism.

Original post →

More from Companies & People

Companies & People channel →