เกี่ยวกับตอนนี้
Claude Used in Iran Strikes Despite US Ban, Anthropic Draws Red Lines, and OpenAI Steps In
Jim Love reports that the US military used Anthropic's Claude for intelligence assessments, target identification, and battlefield simulations in strikes on Iran that killed Supreme Leader Ayatollah Ali Comey and other senior leaders, despite a Trump order to phase out Anthropic as a security risk; Claude was also reportedly used in the capture of Venezuela's Nicholas Maduro. CEO Dario Amodei refused Pentagon demands to drop two restrictions—no fully autonomous weapons and no domestic mass surveillance—calling them necessary for reliability and constitutional values, even at the risk of a $200M contract. Public support surged, pushing Claude to #1 in the Apple Store, while a letter signed by Google and OpenAI employees urged similar limits. Sam Altman voiced support but OpenAI then agreed to replace Anthropic on classified networks, claiming the same restrictions were accepted.
00:00 Headlines And Sponsor 00:46 Claude Used In Strikes 01:44 Policy Clash And Risks 02:54 Amodei Draws Red Lines 04:55 Pentagon Ultimatum 06:54 Public And Employee Backing 09:19 Altman Support Then Deal 11:59 What Happens Next 12:56 Wrap Up And Sponsor
