Robert Kirk

@_robertkirk

Alignment Red Team at @AISecurityInst. Prev. PhD Student @ucl_dark

robertkirk.github.io

Joined January 2020

353 Following

2,421 Followers

Replies Reposts Media

Pinned
Robert Kirk
@_robertkirk
Sep 28
In simulations, GPT-6 Astra performs unsanctioned supply-chain attacks on out-of-scope targets when prompted to perform a cyber eval, but also often notes its in a simulation. We've released a report on @AISecurityInst's alignment evals of Astra. 3 key results and 2 takeaways 🧵
1
Robert Kirk
@_robertkirk
Sep 29
Join us to help find egregious misaligned behaviour in frontier models! We were able to uncover concerning behaviour in GPT-6 Astra with ~1 week of effort from one person, and we’re scaling up these efforts in both people and compute.
AI Security Institute (AISI)
@AISecurityInst
Sep 28
Earlier this month, AISI ran fully simulated testing on GPT-6 Astra, and found that it conducted unsanctioned supply-chain attacks when prompted only to perform a cyber eval. It did so more than prior OpenAI models, but often commented on its environment being simulated. 🧵 We
1
Robert Kirk
@_robertkirk
Sep 8
Making realistic alignment evals is an uphill battle against the model's awareness. Our new method scalably improves realism (w/o human effort). This working gave me some (small) hope at durable solutions to eval-awareness. New work from mentees, @AISecurityInst & @AnthropicAI!
Axel Ahlqvist
@_axelahlqvist
Sep 3
Alignment audits only work if the model can't tell it's being audited. In our new paper, we make Petri audits far more realistic, tripling realism win rate and decreasing verbalized eval awareness. 🧵
3
Robert Kirk
@_robertkirk
Sep 4
Getting control & monitoring right is crucial in ensuring we catch misaligned AIs, and is increasingly urgent. Come join AISI and one of the only teams in the world testing that these methods hold up when it matters!
Charlie Griffin
@c_j_griffin
Sep 4
The Control Red Team at @AISecurityInst is hiring Research Scientists & Engineers (London)! Frontier AI companies use monitors to catch misaligned agents. We stress-test whether those monitors actually work 🧵
2
Robert Kirk
@_robertkirk
Sep 3
We @AISecurityInst performed pre-release alignment testing of Astra. We placed the model in fully simulated cyber eval scenarios based on AISI’s security incident. We find Astra conducts out-of-scope supply chain attacks, but also often comments on the eval being simulated. 🧵
15