Biological weapons: The weaponization of disease

Artificial intelligence could potentially be misused to develop biological weapons, a scenario that scientists, security services, and international organizations are actively examining. Pathogens can now be analyzed and altered with ever greater precision thanks to ongoing developments in biotechnology, while AI can evaluate vast quantities of data at astonishing speeds. Experts warn that modern technologies could be misused to develop biological weapons in the long term.

Security services are particularly concerned about this emerging threat. In October 2026, Ken McCallum, head of the British domestic intelligence service MI5, warned that artificial intelligence could increase the risk from existing, or new, biological weapons – including from non-state actors who might have easier access to these technologies.

Research and misuse dilemma

A report by the American AI company Anthropic in September 2026 highlights the topicality of this debate. Anthropic said it had identified and blocked numerous attempts to use its AI assistant Claude in connection with possible biological weapon research. These attempts included grant applications for so-called "gain-of-function" research, as well as projects to research avian influenza variants that were potentially the most transmissible to mammals.

Although it could have a military application, such research could also relate to medical prevention.

The difficulty in distinguishing between legitimate science and its possible misuse is one of the biggest problems in terms of controlling modern biotechnologies. Many pathogens have the potential for weaponization, with anthrax being one of the most notorious biological agents due to its resistant spores that can survive a long time in the environment.

International control challenges

Biological weapons are prohibited under international law through the Biological Weapons Convention (BWC), which came into force in 1975 and has been ratified by 189 countries. The Convention prohibits the development, stockpiling, acquisition, and transfer of biological and toxin-based weapons. However, unlike chemical weapons, there is no comparable inspection system for biological weapons.

States fear that inspection could expose sensitive research data pertaining to medicine, biotechnology, or the pharmaceutical industry. This is why the control of biological weapons relies to a great extent on transparency, voluntary reporting, and good faith. The 2001 anthrax attacks in the United States, which killed five people, demonstrated the limitations of current control measures.

Detection difficulties

Finding biological weapons programs is much harder than controlling nuclear or chemical weapons production. Many technologies used to develop biological weapons also have legitimate civilian applications, creating a "dual use" problem. Furthermore, a biological attack can initially look like a natural disease outbreak, making it difficult to determine whether a pathogen was deliberately released.

The combination of advancing AI and biotechnologies creates new challenges for international security. As these technologies become more accessible, the international community must develop better methods to prevent their misuse while preserving their benefits for medical research and public health.