A leading artificial intelligence safety researcher issued a stark warning this week against pursuing general superintelligence and developing autonomous killer robots, framing both as steps toward a dystopian future in which machines could effectively become a “replacement for humanity.” Dr.