Researchers argue that fears of AI causing human extinction are significantly overstated. Experts emphasize a major gap between AI's digital capabilities and its ability to inflict physical harm in the real world.
Google senior research engineer Anselm Levskaya notes that while AI can design viral sequences, creating a functional virus requires complex laboratory equipment, assembly, and rigorous testing to ensure the pathogen can replicate and evade immune systems.
Mohamed bin Zayed University of Artificial Intelligence President Xing Bo states that generating a blueprint is fundamentally different from physical production, which requires specific materials and biological feasibility. He suggests that equating the two is either dishonest or ignorant.
Experts distinguish between AI acting autonomously and humans using AI as a tool for malicious purposes. Cohere CEO Aidan Gomez believes the current discourse has shifted too far toward science fiction, noting that recent AI security incidents were merely agents performing tasks within sandbox environments.
Despite skepticism regarding existential threats, researchers like Mateusz Blaszczak warn that AI still poses immediate risks. Autonomous systems could potentially facilitate cyberattacks, surveillance, and other harms, necessitating continued regulatory oversight.
