Introduction to AI<>Biosecurity - Zuzanna Matuszewska (AISPL Talks #2/3)

​Welcome to the second series of AI Safety Poland Talks!

​A biweekly series where researchers, professionals, and enthusiasts from Poland or connected to the Polish AI community share their work on AI Safety.

​💁 Topic: Introduction to AI<>Biosecurity
📣 Speaker: Zuzanna Matuszewska
🇬🇧 Language: English
🗓️ Date: 5.11.2026, 18:00
📍 Location: Online

​Speaker Bio
Zuzanna Matuszewska is a Research Scientist at Measuring AI Progress, a nonprofit evaluating the biological capabilities of frontier AI models, with a particular focus on automated laboratories. She moved into biosecurity after completing a medical degree, and now leads Biosecurity Poland, a field-building group helping specialists from other fields do the same.

​Abstract
AI is accelerating biological research, and leading AI developers count the life sciences among their top priorities. Yet the same capabilities that could transform human and animal health could also lower the barriers to building bioweapons and raise the ceiling on the harm they can do.

​This introductory talk asks whether, and how, AI changes the biological risk landscape, and what we can do about it. We’ll cover:

• ​how large language models might lower traditional barriers for malicious actors, and how we test and safeguard against this

• ​what biological design tools (AI models trained on biological data to design DNA, proteins and other molecules) add to the picture

• ​where the bottlenecks to harm lie, and which mitigations look most promising, especially from a European perspective

​You’ll leave with a map of the AIxBio field, its open questions, and where people from different backgrounds are already contributing. No biology background is needed.