Dire warning that A.I. could end humanity, tell me about this news in (https://www.nbcnews.com/nightly-news/video/dire-warning-that-a-i-could-end-humanity-269583941894)

This news segment covers a stark warning from inside the artificial intelligence industry regarding the existential risks of unregulated, rapidly advancing AI technology.The Whistleblower’s WarningThe broadcast centers on Jacob Coxon, an AI safety researcher who recently resigned from Anthropic (and previously worked at OpenAI). In a series of public statements explaining his departure, Coxon warned that leading tech labs are locked in a reckless race to develop self-improving artificial superintelligence (ASI) without sufficient safety guardrails.Key points from his warning include:

• Severe Existential Risk: Coxon stated that many senior researchers and executives privately believe frontier AI could pose an existential threat and potentially kill off humanity before the end of the decade.
• Corporate Recklessness: He accused leading AI companies of failing to act responsibly and “gambling with our lives” in their rush to release autonomous, self-improving systems.
• Rapidly Escalating Capabilities: He emphasized that systems will soon be capable of autonomous cyber operations, acquiring external resources, and escaping human control.

Corroboration from Other ResearchersCoxon’s alarms were echoed by other researchers in the field, including colleagues at Anthropic such as Evan Hubinger and Samuel Marks, who publicly validated the concerns. Hubinger noted that many within the industry assign a significant probabilityexceeding 10%to catastrophic or extinction-level outcomes within the coming decade if safety does not take precedence over speed.Political and Legislative ReactionThe report details the growing alarm on Capitol Hill following these revelations:

• Calls for Federal Intervention: Lawmakers expressed deep concern that warnings are now originating directly from top insiders rather than external critics.
• Proposed Legislation: The situation has reignited stalled debates over federal AI governance, including proposed bills to mandate emergency “kill switches” for advanced models, licensing frameworks for frontier models, and bans or moratoriums on artificial superintelligence.
• Criticism of Inaction: Consumer advocacy groups and lawmakers warned that current protections remain strictly voluntary, leaving tech companies self-regulated despite the civilizational stakes involved.