Test a browser agent on web games
Researchers had ChatGPT Atlas's browser agent play T-Rex Runner, Sudoku, Flappy Bird and Stein.world; it did well on logic puzzles like Sudoku, finishing faster than humans.
What they did
On each game's page in the ChatGPT Atlas browser, the researchers turned on Agent Mode (Preview) and gave one zero-shot instruction: play until you get stuck. They ran 10 trials per game and did not intervene. The agent worked out the controls itself. For Stein.world, they later added a second prompt that spelled out the keys and the first task.
How it went
Sudoku took an average of 2m28s with 100% accuracy, against a human baseline of 10-12 minutes. T-Rex averaged 45.5 and Flappy Bird scored 0 in every trial. In Stein.world it only got out of the first room once it was given the detailed prompt.
Worth knowing
The sample is small and the human baselines are rough, so the authors call the results preliminary. Stein.world needed explicit instructions, and even then the agent spent a long time deliberating before each move.
Labels
Researchers had ChatGPT Atlas's browser agent play T-Rex Runner, Sudoku, Flappy Bird and Stein.world
They ran 10 trials per game and did not intervene. The agent worked out the controls itself.
The agent worked out the controls itself.
On each game's page in the ChatGPT Atlas browser, the researchers turned on Agent Mode (Preview) and gave one zero-shot instruction: play until you get stuck.
They ran 10 trials per game and did not intervene.
The “Material” label failed its quality check and is not used for selection.
Details
- Agents
- ChatGPT
- Source
- other
- Source date
- undated
- Accounts
- 1
- In registry since
- Oct 12, 2026