North East Regional Cyber Security Seminar Series - Who Is Really Answering? Fingerprinting and Stress-Testing Large Language Models from the Outside

Organisations increasingly build products on large language models they can only reach through an API, with little means of checking which model is actually answering or how it behaves under attack. Dr Eaby Kollonoor Babu will speak on auditing LLMs from the outside. A model leaves measurable traces in the way it responds: the timing of its token stream alone can reveal the provider, and in many cases the specific model, without reading a word of the output. It can also be examined how much behavioural probing can, and cannot, reveal about a model's underlying architecture. The discussion will then turn to security, testing how well leading commercial models hold their instruction hierarchy against multi-turn prompt-injection attacks hidden in untrusted content, and whether they recover once compromised. The talk will close with the implications for model verification and AI supply-chain assurance. The session includes a live demo, and it’s followed by a networking lunch. You can join in person or online.