[Job] Research Scientist / Engineer opportunity in UK AISI's Criminal Misuse team
Kimberly Mai
Hi everyone,
I'm reaching out from the UK AI Security Institute. We're working to get governments to deeply understand frontier AI risks and to advance mitigations research, and we have ~70 technical staff, many of whom joined from DeepMind, OpenAI, and Anthropic.
I’m a researcher in the Criminal Misuse team at AISI, led by Prof Christopher Summerfield. The team focuses on investigating how frontier AI systems could be misused for fraud, social engineering and radicalisation. We’re looking to recruit more researchers (JD here).
The role is hugely varied and can involve:
- Developing threat models of how adversaries could use frontier AI to support criminal activity
- Running evaluations, red-teaming, and demos on AI-enabled fraud, social engineering, and radicalisation
- Supporting AI developers with insights around gaps in refusal training
We’re looking for
- A strong ML/AI research background (e.g. PhD but not essential)
- LLM knowledge and technical safety expertise
- Mission-driven mindset with excellent communication skills
- Interest in preventing AI misuse
It’s a great role that offers the opportunity to conduct technical AI research that translates into direct policy impact. Happy to share more details if you are interested, and happy to chat about my experience as well.
Best,
Kimberly Mai