Why AI Isn’t Likely to Wipe Out Humanity With Bioweapons
wired ·
Despite rising fears about an AI-driven apocalypse, scientists interviewed by WIRED say that AI-caused human extinction through bioweapons is currently a relatively low-probability threat. AI can make biological information and protocols easier to find and use, but experts say the harder bottlenecks are obtaining materials and equipment, assembling and testing a pathogen, and demonstrating human infectivity and transmission. Biotech and immunology experts argue that human oversight, limited lab autonomy, and potentially rapid AI-assisted vaccine development would make an autonomous attack difficult. Other researchers warn that malicious people could use AI to arrange bioterrorism even before fully autonomous labs exist, and that future capabilities could increase the danger. Some experts also say pathogens are poor weapons because they are hard to target and mass-produce compared with conventional attacks. Proposed safeguards include screening synthetic DNA and RNA orders, strengthening AI-model protections, improving outbreak surveillance and data sharing, and adding broader biosecurity measures. Experts emphasize layered defenses while also warning that AI-safety debates should not obscure the technology’s potential benefits for vaccines and medicine.
wired