An Anthropic co-founder said AI systems should be required to include a “kill switch,” according to a just-in report. The statement highlights ongoing concerns around AI control and safety as the technology continues to advance.