Could AI really kill us all? Your questions, answered.

MIT Technology Review examines whether AI could cause human extinction and concludes that such an outcome is highly unlikely. However, autonomous weapons, cyberattacks, biological threats, unreliable agents, weak monitoring, and limited regulation already pose serious risks.

《麻省理工科技评论》探讨了人工智能是否可能导致人类灭绝，并认为这种结果极不可能发生。然而，自主武器、网络攻击、生物威胁、不可靠的智能体、薄弱的监控以及有限的监管已经构成严重风险。