cd /news/artificial-intelligence/an-ai-scientist-that-doesn-t-drift-t… · home topics artificial-intelligence article
[ARTICLE · art-91431] src=arxiv.org ↗ pub= topic=artificial-intelligence verified=true sentiment=· neutral

An AI Scientist that Doesn't Drift: Taste, Structure, and Falsifiable Findings in a Quadruped Navigation Research Loop

A new AI Scientist research loop, built on Karpathy's autoresearch paradigm, prevents drift in quadruped robot navigation experiments by using an immutable experiment card, specialized subagents, and a preference oracle called kkanbu. In tests across eleven research streams, both arms falsified about three quarters of their hypotheses, with the oracle-less arm producing the best trained policy, while kkanbu uniquely drove exploration of test-time adaptation and cross-stream learning.

read1 min views1 publishedAug 11, 2026

arXiv:2608.07542v1 Announce Type: new Abstract: Autonomous research loops driven by large language models can run machine-learning experiments at scale but tend to drift toward local refinements of whichever metric they optimise rather than testing the hypotheses that motivate the experiments. We address this structurally and present an AI Scientist for studying generalisation in quadruped robot navigation policies in simulation. Building on the autoresearch paradigm of Karpathy, our loop adds three components: an immutable experiment card that pairs each iteration's prediction with its outcome under a fixed schema, so a falsified hypothesis cannot be retconned; specialised subagents restricted to mechanical roles; and kkanbu, a preference oracle that holds the user's research taste as a typed knowledge graph and is the only component permitted to make subjective judgements. To isolate the oracle we run the identical loop twice across eleven research streams, with and without kkanbu. Neither arm drifts: both falsify roughly three quarters of their own hypotheses, and the best trained policy comes from the oracle-less arm. What the oracle changes is direction, not score: it alone explores test-time adaptation, it authored the winning designs where its arm led, and it carried lessons across streams that the other arm repeatedly re-derived. The scaffold keeps the loop honest; kkanbu decides where it looks.

── more in #artificial-intelligence 4 stories · sorted by recency
── more on @arxiv 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/an-ai-scientist-that…] indexed:0 read:1min 2026-08-11 ·