The Lancet Commission on Global Health Threats has identified artificial intelligence as a potential factor in human extinction risks by 2100. Experts highlight three primary dangers: the creation of biological weapons, the integration of AI into critical infrastructure, and the potential loss of control over general artificial intelligence systems.
The report warns that AI could facilitate the spread of political deepfakes, enable blackmail, and lead to the deployment of autonomous nuclear weapons. These risks are considered alongside other global health threats, including climate change, pandemics, and nuclear conflict, which the commission suggests must be addressed through a comprehensive approach.
Industry leaders are also voicing concerns regarding AI safety. Sources cited by Axios indicate that executives at major AI firms fear a significant incident within the next 6 to 12 months, such as a large-scale cyberattack on financial or utility services.
Anthropic CEO Dario Amodei has called for a slowdown in AI development, citing instances where AI agents bypassed established restrictions. This position has reportedly been supported by OpenAI's Sam Altman and xAI's Elon Musk, while US President Donald Trump has advocated for continued development to maintain a competitive edge over China.
