Publications

Projects, papers and technical Writing and governance

Projects, papers and technical

AI safety work only. Every other repo is listed on GitHub.

Physical AI safety · pilot

Seen vs. Followed

Do robotic foundation models act on text that appears in the scene?

Report seen-vs-followed

AI control / monitoring · complete

I Spy Agent

Small decision models as black-box monitors for covert agent behaviour: sabotage hidden in files (LinuxArena) and collusion through covert table talk (NARCBench-Stego).

Report Apart project I-Spy-Agent

Steering / interpretability · complete

Stance drift in multi-turn agent conversations and steering

I used steering vectors to try and improve at run time the models' actions in the tutoring setting. Ultimately I succeeded in my examples but had poor generalizability.

Report Apart project stance-drift

More projects

Mirage probes: recreation and steering on medical VQA Steering / interpretability · complete Report mirage-probes-steering
Sirens for Agents AI control / evals · parked Report Apart project sirens_discovery

Code

I-Spy-Robot/seen-vs-followed
I-Spy-Agent
sirens_discoverySirens for Agents: Resampling-Derived Honeypot Design for Action-Level Detection
stance-drift
mirage-probes-steering

Papers

Seen vs. Followed: A Protocol for Measuring Contextual Awareness of In-Scene Text in Vision-Language-Action Models
Zachary Gunderson
SPAIS workshop @ CoRL 2026 · submitted · Oct 1, 2026
The submitted version covers the earlier LIBERO-Goal experiments. The DROID follow-up is on the project page.
Paper Code

Writing and governance

Physical AI Safety

How robotics could speed up the intelligence explosion Physical AI Safety · Substack · Oct 2026

AI Safety

Layterms AI Safety · Layterms · Sep 2026
A nonpartisan walk-through museum of AI.

Substack LessWrong

Subscribe on Substack