Skip to content
forward pass

Claude learns to use a computer

Anthropic releases a model that looks at a screen, moves a cursor and types; the assistant becomes an agent, and the year of agents that follows starts here.

category
product
significance
3 of 5
people
Dario Amodei
organisations
Anthropic

what had to happen · 54 events back to 1943

Every event this one built on, transitively, in order. The organism has the same path lit. Direct influences are marked.

00 · One neuron · 1

  1. 1943A logical calculus of nervous activity

I · Foundations · 9

  1. 1948A mathematical theory of communication
  2. 1949Cells that fire together wire together
  3. 1950Programming a computer for playing chess
  4. 1958The perceptron learns
  5. 1959Samuel's checkers program coins 'machine learning'
  6. 1960ADALINE and the least-mean-squares rule
  7. 1965Moore's law
  8. 1969Perceptrons
  9. 1970Reverse-mode automatic differentiation

W1 · The first winter · 2

  1. 1974Werbos applies backpropagation to neural networks
  2. 1980The Neocognitron

II · Connection · 3

  1. 1982The Hopfield network
  2. 1985The Boltzmann machine
  3. 1986Backpropagation

W2 · The second winter · 6

  1. 1988Temporal-difference learning
  2. 1989Q-learning
  3. 1989LeNet reads handwritten postcodes
  4. 1990Finding structure in time
  5. 1991The vanishing gradient problem
  6. 1992TD-Gammon reaches world-class backgammon

III · Statistics and data · 10

  1. 1997Long short-term memory
  2. 1998MNIST and LeNet-5
  3. 1999The first GPU
  4. 2003A neural probabilistic language model
  5. 2006Deep belief networks and the word 'deep'
  6. 2007CUDA
  7. 2009Deep learning moves to GPUs
  8. 2009ImageNet
  9. 2010Rectified linear units
  10. 2010DeepMind is founded

IV · Deep learning · 11

  1. 2012Google Brain's network discovers cats
  2. 2012Dropout
  3. 2012AlexNet wins ImageNet
  4. 2013Deep Q-networks play Atari
  5. 2014Google buys DeepMind
  6. 2014Superintelligence
  7. 2014Attention
  8. 2014Sequence to sequence learning
  9. 2015Batch normalisation
  10. 2015Residual networks
  11. 2015OpenAI is founded

V · Transformers · 9

  1. 2017Attention is all you need
  2. 2017Deep reinforcement learning from human preferences
  3. 2018GPT: generative pre-training
  4. 2019GPT-2 and the model too dangerous to release
  5. 2020Scaling laws for neural language models
  6. 2020GPT-3
  7. 2020Learning to summarise from human feedback
  8. 2021Anthropic is founded
  9. 2021GitHub Copilot writes codedirect

VI · Everyone · 3

  1. 2022InstructGPT
  2. 2023Claude
  3. 2024Claude 3 catches GPT-4direct

On 22 October 2024 Anthropic released an upgraded Claude 3.5 Sonnet with a capability it called computer use, in public beta. Given screenshots, the model could work out where things were on the screen, move a cursor, click, scroll and type, and so operate any software written for people: fill in a form from a spreadsheet, look something up in a browser, run a command. It was slow, made mistakes a person would not, and on the company's own benchmark completed about fifteen percent of the office tasks it was set. It was also the first general-purpose agent from a frontier laboratory that acted rather than advised.

The release framed the next eighteen months. OpenAI's Operator followed in January 2025, Google's Project Mariner and the browser agents of 2025 after it, and the coding agents, Claude Code and Codex, that by 2026 wrote a large share of the world's new software began as this idea applied to a terminal.

Computer use is on this timeline as the point at which the models got hands. The organism on this page, in the final chapters, is wired to tools for the same reason.

what it led to · 5 events downstream, through 2026

Built on it directly:

  1. 2024The Model Context ProtocolVI
  2. 2025Claude 4 and Claude CodeVII

And, through them, by era:

sources · 1

Trace the lineage of this event on the timeline →Back to the ledger