Exploring AI-Biosecurity Governance in the Global South

Artificial intelligence (AI) has the potential to advance life sciences, enabling breakthroughs in personalized medicine and public health. However, these advancements also present serious dual-use concerns, particularly the potential for AI to facilitate the development and deployment of biological weapons.

The COVID-19 pandemic has demonstrated how unprepared the world is to deal with a major biological threat. This vulnerability has been continually reflected in the Global Health Security Index. While no country is fully prepared, gaps in healthcare infrastructure and biosecurity capabilities among Global South countries make these regions particularly vulnerable to biological threats, both natural and man-made.

For a deeper dive into this topic, read the full version here.

Unpacking the Risk

AI models with biological capabilities are advancing rapidly. Experts are increasingly worried that AI could facilitate the development of biological weapons, as reflected in major policy documents such as the White House Executive Order on AI, the Bletchley Declaration, and numerous papers and think tank reports. While promising for research, these technologies could alter the threat landscape of biological weapon development in two main ways:

1. AI may lower existing barriers to biological weapon development, increasing the number of potentially capable actors.
2. AI developments could increase the potential harm of a biological weapon by facilitating the design of modified or novel biological agents with enhanced disease characteristics.

Insufficient Regulation

Despite the catastrophic potential of AI misuse for biological weapon development, only 27% of the 141 Global South countries assessed in this research have national AI strategies, none of which address biosecurity concerns. Effective regulation is crucial to mitigate adverse incidents, which could not only lead to loss of lives and destabilization but also undermine public and investor confidence in emerging bioeconomies in these regions, which are vital for economic growth and public health. Regional frameworks like the African Union’s Draft Policy and ASEAN’s Guide on AI Governance and Ethics offer starting points for regulation but omit AI–biosecurity risks.

Meanwhile, the Global North, where most AI models are developed, has only taken initial steps toward addressing these risks. Current AI strategies, if they mention biosecurity risks at all, primarily discuss reporting requirements and intentions to do risk assessments, and most of them exclude small and narrow AI models. However, the level of computational resources used to train an AI model does not necessarily correlate with its potential biosecurity risk, meaning that even smaller models can pose considerable threats.

Participation in Global Dialogues

The Global South must actively engage in global dialogues on regulating AI models, advocating for robust regulations in the Global North to address dual-use risks while ensuring these do not hinder AI’s potential to improve healthcare and pandemic preparedness. Given their distinct healthcare challenges and technological needs, Global South countries bring unique perspectives to these discussions, which can help create more inclusive and effective governance frameworks.

Recommendations

1. Integrate AI-biosecurity into strategies.
   Countries and regional organizations like the African Union and ASEAN should include AI-biosecurity challenges in their AI strategies and biosecurity policies to enable governments to move towards legislation which mitigates risks at the intersection of AI and biosecurity.
2. Push for international regulation.
   By engaging in global dialogues, such as NTI’s AI-Bio Global Forum or the AI Action Summit in France, the Global South can advocate for risk assessments and comprehensive regulations in countries where AI models are developed. Active participation ensures global policies reflect diverse needs and address risks that disproportionately impact the Global South.
3. Strengthen biosecurity defenses and mitigation capabilities.
   Enhancing biosecurity measures, such as synthetic DNA screening, can raise barriers against the successful misuse of AI in biological weapon development, while investments in public health infrastructure can improve the Global South’s capacity to effectively respond to biological threats.

AI offers transformative opportunities for the Global South, but its potential risks must be managed with foresight and collaboration. By integrating AI-biosecurity into policies, advocating internationally for regulation, and enhancing local biosecurity capabilities, the Global South can protect its communities while contributing to global security. Although most AI models are developed in the Global North, the Global South’s perspectives are critical to crafting inclusive, effective policies.