# Artificial Atlas

> A sourced map of the research camps in AI: who is betting on what, against whom, and why. Six camps defined by four fault lines; dated roles, funding rounds and camp assignments carry a URL and date. Data is CC BY-SA 4.0. Last changed 2026-09-05.

Cite as: Artificial Atlas, artificialatlas.org, retrieved on the date of use. Person entries hold only public professional facts.

## Start here

- [Start](/start/): the six camps in plain words and where the money went.
- [Fault lines](/faultlines/): the four questions the camps answer differently.
- [Capital](/capital/): every round since 2022 by camp, with the case against.
- [People](/people/): everyone in the atlas by camp.
- [API](/api/): the whole dataset as JSON.

## Camps

- [Scale next-token prediction](/camp/llm/): Internet-scale text pretraining, then reinforcement learning on outcomes, then tools. Grounding, planning and memory are expected to emerge from scale.
- [World models](/camp/wm/): Predict the world, not text. Learn the dynamics of the environment from video and sensors so an agent can imagine outcomes and plan.
- [Physics AI and AI for science](/camp/phys/): Ideas are cheap now; experiments are the bottleneck. Replace the lab with a universal learned simulator that also tells you which direction to improve.
- [Learn from experience](/camp/rl/): Intelligence is online reinforcement learning from a reward stream. Static datasets and the train-then-deploy split are a detour, and current deep learning cannot learn continually.
- [Physical AI and robotics](/camp/robot/): The architecture is roughly settled; data is the bottleneck. Scale vision-language-action models on teleoperation, human video and simulation until robots generalize.
- [Program search and other substrates](/camp/alt/): Gradient descent over one giant network is the wrong primitive. Intelligence is efficient skill acquisition on novel problems, which needs search, symbols, probabilistic models, or a different kind of computer.

## Data

- [people.json](/api/people.json): people with dated roles, camps and sources
- [labs.json](/api/labs.json): labs with camp memberships and funding rounds
- [rounds.json](/api/rounds.json): every round
- [contrarians.json](/api/contrarians.json): positions against the AI trade
- [feed.xml](/feed.xml): moves and data changes

## Full text

- [llms-full.txt](/llms-full.txt): camps, labs and written person entries in one file