We ran AI agent swarms in Minecraft and Mindustry
READ THE ORIGINAL SOURCEOpen original sourcelabs.ycombinator.com · Opens in a new tabThe authors ran long-running LLM agents in Minecraft, including a 50-agent survival run for three hours with the goal of building a thriving civilization. The agents spread out, quickly harvested nearby trees, and formed a hierarchy around the first agent in the directory order. Agents coordinated through shared plans, a ration ledger, and direct messages. Most agent deaths occurred on the first night, after which agents became more risk-averse and focused on food and survival. The authors open-sourced their agent harness.
One primary source; other links are context, not independent corroboration.
- Primary sourceHacker News · AI
- Discovery linkHacker News
- Linked in sourceFollow on Twitch
- Linked in sourceOlam Labs
- Linked in sourcevideo pretraining work
- Linked in sourcedefeating the Ender Dragon
- Linked in sourceMindustry
- Linked in sourcethe OpenAI and Hugging Face incident
- Linked in sourceMulti-Agent Mindustry
- Linked in sourceMulti-Agent Harbor
- Linked in sourceHarbor
A community post points to the original link. Discussion and upvotes do not establish accuracy.
Open full article view →- Community discussion+5
- Fetched source page+30
- No editor link review+0
- Community link+0
This measures the stored source trail and editorial steps. It does not certify accuracy, peer review, or independent corroboration. Popularity adds no points.
View full scoring →Read the scoring rubric →What the source says
The authors ran long-running LLM agents in Minecraft, including a 50-agent survival run for three hours with the goal of building a thriving civilization. The agents spread out, quickly harvested nearby trees, and formed a hierarchy around the first agent in the directory order. Agents coordinated through shared plans, a ration ledger, and direct messages. Most agent deaths occurred on the first night, after which agents became more risk-averse and focused on food and survival. The authors open-sourced their agent harness.
A READER'S EVIDENCE CHECKHow to read this signalSix questions about claims, sources, risks, and open checks
This is a signal-level guide until claim-level evidence is available. “Not recorded” means the evidence has not been assessed here; it does not mean a result is false. How we check AI news →
- 01Who is making each claim?
The claimant and whether each claim was measured or asserted are not recorded in this public brief.
- 02What was actually tested?
Not recorded per claim. Check the source for sample size, tasks, model and benchmark versions, and reproducibility.
- 03Where are the sources?
Original link: labs.ycombinator.com. Brief basis: fetched source page. Primary page date and quote locations are not established in this panel.
- 04What could mislead?
Source claims not independently checked; the study is a small, single-environment experiment and the results may not generalize.
- 05What is at risk if wrong?
Not assessed per claim. Consider safety, privacy, cost, and adoption decisions before relying on the result.
- 06What would change the view?
No claim-specific check is recorded. An inspectable primary method or independent replication could change the assessment.
Key findings from the stored brief
- Fifty agents spent the run in shelters rationing scarce food and built a pseudo-government with a ration ledger and rules.
- The agents sent about 5,900 messages and an agent directory order produced a de facto leader, Ada.
- 22 of the run's 34 agent deaths occurred in the first night, after which agents became increasingly risk-averse.
- The authors open-sourced a Minecraft agent harness built on Pi and Mineflayer tools.
Source claims not independently checked; the study is a small, single-environment experiment and the results may not generalize.
This is a source-based abstract, not independent validation of the source's claims.
Evidence & sources
These links show where the signal came from and what the source page linked to. A linked page is context, not independent corroboration.
Related archive coverage
These relationships record a citation or topic overlap, not confirmation of the finding.
A community post points to the original link. Discussion and upvotes do not establish accuracy.
Timeline & updates
- Signal dated by source or discovery feed
This timestamp places the signal in the radar. It may be the discovery post date rather than the primary page's first publication date.
- Shared brief stored
AINewsIQ prepared one abstract from fetched source page.
A publication revision history is not stored for this signal. New engagement snapshots do not mean the source itself changed.
Why this may matter
The experiment offers concrete observations of how large numbers of language-model agents form hierarchies, share resources, and adapt behavior under survival pressure, which is relevant to multi-agent alignment research.
Limits and open questions
Source claims not independently checked; the study is a small, single-environment experiment and the results may not generalize.
Organization-specific financial, security, and operational effects require the organization's own context and review.
Discussion
Comments are reviewed before publication. Discuss the evidence, not the person.
No published comments yet.
Reader discussion opens with accounts. You can read the original source and published comments now.