Robust generalisation

Doing as well on messy real work as on benchmarks.

Facts

Lane: Algorithms and learning
Wave: 6 of 12 (build order, not a date)
Needs: 0 boxes all the way back through hard links
Unlocks: 3 boxes downstream through hard links
Leverage: 0.5 points · stands in the way of 2 of 53 stages · rank 17 of 57
Cluster: Agents at work

Evidence as of October 2026

Sutskever says the field is moving from an age of scaling to an age of research, because people learn from far less data than models need and AI must learn as efficiently.

01Sutskever: a new learning paradigm is necessary

Around it

Needs · 0

Nothing beyond the detail list. It can start at any time.

Unlocks · 3

AGI: top-expert-level AIstage · Speculative
Continual-learning agentsstage · Emerging · helps, not required
Agent-run discoverystage · Emerging · helps, not required

Loops and off-ramps

Capability plateauoff-ramp · hit directly

Openings researchers' judgement, October 2026 scouting

Startup room 27/100 · rank 116 of 124 Research room 62/100 · rank 23 of 124

Crowding4 Lab overlap5 Small-team fit2 Buyer today2 Research openness5

Depth: deep
Players: SSI (~$8bn); Ineffable Intelligence ($1.1bn seed); AMI Labs (~$1.03bn); Adaption Labs; every frontier lab
Wedge: Measurement, not paradigms: private, refreshed real-work suites that quantify the benchmark-to-reality gap and domain generalisation audits before deployment. research