Even Without AI, the Risk of a Biological Attack Is Terrifying (www.motherjones.com)

Recent discussions surrounding AI have sparked fears of existential threats, but a chilling report from Anthropic emphasizes a more immediate danger: biological attacks. The report revealed attempts by foreign scientists to bypass AI tools like Claude to conduct "gain of function" research, which alters pathogens to make them more virulent. Such modifications, while potentially useful for vaccine development, also pose significant risks if misused. Anthropic responded by blocking these users, underlining the dual-use dilemma that accompanies advanced AI and genetic engineering technologies. The significance of this narrative extends beyond hypothetical threats; it highlights the alarming accessibility of tools that can enable biological warfare. With thousands of high-security labs globally conducting risk-laden research, the potential for accidental release of engineered pathogens is notably greater than deliberate attacks by rogue entities. Experts like Annie Jacobsen stress that the combination of advanced genetic engineering and AI capabilities, though not explicitly addressed in her latest book, exacerbates the risks of creating deadly airborne agents. As the public grows increasingly aware of these dangers, the AI and ML community must navigate the ethical landscape of technology’s dual-use potential.