Skip to the card

2026-11-17 issue

Observed arrival · 2026-11-17

Humanlike’s home-robot safety test

humanlikelabs.tech Visit website
Editorial interest 78/100 Selection signal · not a rating of the site

Humanlike presents a research project testing whether fine-tuning an AI model for a job restores hazardous knowledge filtered from its training data, with home robots as the motivating case.

Landing page captured for the 2026-11-17 issue.
For
People researching AI safety for home robots
Worth noticing
The page separates job-test gains from format effects and says no deliberate attack has yet been run.

Field notes

The experiment uses three open Pythia models drawn from EleutherAI's Deep Ignorance research: an unfiltered model, a model with hazardous biology filtered from training, and a filtered model with additional safeguards. Each received the same 3,476 examples for a cosmetics-lab-assistant task. Humanlike reports a 150-question held-out job test and WMDP-Bio scores, while noting format effects, the limited model size, and that deliberate attacks remain future work.

Observed signals

Read the marks

Editorial observations of this landing page, not a rating.

○OpenPublic substance visible
✦PrettyNotable craft visible
●ProPolished or operationally mature
◎NicheUnusually specific use

One card from the complete issue

Real Weather in Mossy Hollow

300,778 arrived 1,000 judged 976 catalogued Enter the complete issue
humanlikelabs.tech

Landing page observed 2026-11-17. The live site may have changed.