Africa’s AI Capacity Gap Raises Biosecurity Concerns — Ferguson

Professor Neil Ferguson, director of public health at Imperial College London, has expressed concern that Africa may face a growing gap in artificial intelligence (AI) capability. AI is currently concentrated in a limited number of countries, with Africa lacking the necessary data center capacity compared to Europe, the United States, and China. As AI becomes increasingly relevant to healthcare, scientific research, and biosecurity, Ferguson stressed the importance of addressing this capacity gap.

During an online Biosecurity Media Briefing organized by Imperial College London, Ferguson emphasized that while biological threats facing Africa are not fundamentally different from those in other regions, current gaps in scientific knowledge about many viruses pose a limit to the threat AI could pose in the near term. However, there are concerns about AI accelerating automated biological experimentation and helping generate data that is currently missing, which could potentially increase our ability to understand and engineer pathogens.

Plant pathogens are identified as a potentially greater short-term concern, as automated experimentation could generate better data on plant disease, making it easier to engineer threats to crops. The economic consequences of targeting agriculture through malicious means may also provide a clearer incentive for malicious actors compared to creating a pathogen capable of affecting people globally.

Professor Wendy Barclay, Regius Professor of Infectious Disease at Imperial, highlighted the importance of controlling access to synthetic DNA as a short-term biosecurity measure. Although synthetic DNA companies already self-regulate access to potentially dangerous sequences, Barclay argued that stronger international coordination was required to address these concerns.

Lastly, Professor Tom Ellis, a professor of synthetic genome engineering at Imperial, noted that natural evolution remains more capable than current AI systems at exploring biological possibilities. AI-assisted DNA design has not yet reached the point where it can compete with the vast biological search space explored through natural evolution.