Guardian reporters address reader questions on AI doomsday scenarios

The Guardian's technology team—Aisha Down, Blake Montgomery, and Dan Milmo—answered reader questions this week about AI's potential to cause catastrophic harm. One reader asked how AI could realistically kill large populations, questioning whether warnings about bioweapons, nuclear war, or societal collapse were credible.

Down responded that current AI is unlikely to significantly alter the bioweapon equation, noting that historical obstacles to biological attacks stem from logistics rather than invention difficulty. She cited the failure of Aum Shinrikyo's botulism and anthrax dispersal attempts in Tokyo, arguing that chemical weapons already exist without AI advancement and face similar implementation challenges.