cd /news/ai-safety/should-we-be-worried-about-how-good-… · home › topics › ai-safety › article
[ARTICLE · art-72319] src=lesswrong.com ↗ pub= topic=ai-safety verified=true sentiment=· neutral

Should we be worried about how good AI is getting at coding autonomous drones?

Andon Labs introduced Drone-Bench, a benchmark where AI agents code drones to complete autonomous surveillance tasks, based on its Project Pilot work with Anthropic. The benchmark raises questions about the safety implications of increasingly capable AI coding autonomous drones.

read1 min views1 publishedJul 24, 2026
Should we be worried about how good AI is getting at coding autonomous drones?
Image: Lesswrong (auto-discovered)

Hey everyone!

We’re introducing Drone-Bench, a benchmark where AI agents code drones to complete a simple autonomous surveillance task. Drone-Bench is independent but based on Project Pilot, our work with Anthropic.

I would make a condensed version here for LW, but visuals are much better on web:

https://andonlabs.com/evals/drone-bench Very curious on your feedback!

source & further reading
lesswrong.com — original article Coefficient Giving just gave GiveWell $1 billion. Where should other donors give now? LLMs are (still) mostly powered by imitative learning, not RL Democracy isn’t ready for the AI revolution
── more in #ai-safety 4 stories · sorted by recency
── more on @andon labs 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
→ Live at https://your-agent.zahid.host ✓
Get free account → Pricing
from €0/mo · no card required
LIVE [news/should-we-be-worried…] indexed:0 read:1min 2026-07-24 · —