Deep Dive · XiaoHu Brief

Princeton professor at ICML on AI and work: how should individuals adapt?

In ~24 months, model capability rose sharply; SAGE’s composite reliability metric rose only 5–10 percentage points
60-second takeaway
  • Princeton CS professor Arvind Narayanan’s ICML 2026 keynote in Seoul, “What will be left for us to work on?”, builds on the “AI as Normal Technology” framework with Sayash Kapoor.
  • SAGE lab: over ~24 months, frontier models jumped in capability; reliability (consistency / robustness / calibration / operational safety) rose only 5–10pp.
  • In software’s decide–execute–deliver stack, AI mainly compresses execute (~1/3 of effort); decide and deliver are not compressed—and may expand.
  • ATM, radiology, translation, software tools: automation rarely cuts jobs 1:1; software employment grew ~10,000× through many ~10× tool leaps.
  • He splits RSI / human-level AI / economically transformative AI / superintelligence into four non-entailing dimensions.
  • Personal adaptation: raise the ceiling, don’t only ride the floor; balance productivity / growth / control; refuse black boxes; master first, then amplify; reinvest ~10 hours/week into skills.
1Opening

Two narratives: the AI field is anxious about its own jobs

At ICML 2026 in Seoul, Princeton CS professor Arvind Narayanan gave a keynote titled “What will be left for us to work on?”

He faces the anxiety head-on: as AI can do more of what we do, how should we prepare?
🎯
Why it matters: He runs SAGE on AI agent evaluation, co-authored the long essay AI as Normal Technology (~15k words, becoming a book). The talk ties the framework, self-built reliability data, and frontline software observations together.
What will be left for us to work on?
Title slide. Source: Arvind Narayanan · ICML 2026 keynote slides

He splits the path into two practical camps—not pure philosophy:

Narrative 1 · Replacement
In a few years AI replaces almost everything we do today. Rational move: accumulate wealth before skills devalue. Some in Silicon Valley take this path; the “permanent underclass” meme lives outside the lab too.
Narrative 2 · Amplification (his camp)
AI will greatly amplify human potential. Now is the best time to build complementary skills—agency, taste, judgment—and the scaffolding around them.
Two narratives
Replacement vs amplification. Source: keynote slides

If you bet on replacement and amplification wins, you may miss the best window in history to build superpowers. The world is watching how the AI community responds; rolling over and accepting “AI will do the work” may fuel a sharper political backlash.

2Framework

Four stages of AI impact—the slowest has barely begun

“Normal” in AI as Normal Technology does not mean AI is a hammer or a toothbrush. They treat it as industrial-revolution scale, as a causal model of how capability becomes economic and social impact.

Classic diffusion: invention → innovation (appliances) → adoption. They expand it into four stages, with software as the example:

① Methods / capabilityModels improve fast
→
② Products / appsCoding agents, etc.
→
③ Early adoptionVibe coding → agentic engineering
→
④ AdaptationOrg redesign over decades
Four stages
Diffusion applied to AI. Source: keynote slides

Stage four is slowest. Even in software—an early coding-agent adopter—true organizational redesign has barely started. Speculation: if agents can ship huge, secure codebases, one-size-fits-all software for billions makes less sense; software becomes extremely personalized, and even “software company” as a form may be renegotiated. That is human and organizational change—historically measured in decades.

3History