Artificial intelligence can’t be trusted to auto-correct emails, much less make civilization-altering decisions about the use of lethal force, a peace activist told Sputnik on Wednesday. Putting autonomous weapons into high-risk situations only risks triggering a war nobody can back down from.