0:00 Liquid AI just released a 350-million-parameter model that outperforms models twice its size.0:07 This isn't just another benchmark win...0:10 it's a signal that the entire AI race is changing lanes.0:14 For years, the game was scale.0:16 More data, more parameters, bigger models.0:20 But the performance gap between frontier models like GPT-4o and Claude 3.7 has shrunk to just a few percentage points.0:29 The model itself is becoming a commodity.0:32 The new competitive moat isn't the engine...0:35 it's the driver.0:37 It's what the community is calling Agent Experience, or AX.0:41 Users don't care if your model is one percent better on a leaderboard.0:47 They care if the agent understands their intent, handles errors gracefully, and actually gets the job done.0:55 This is why the most talked-about founders aren't building another general-purpose assistant.1:02 They're building what one analyst called "the most boring, domain-specific, deeply embedded software the world has ever seen." Because that’s where a superior agent experience can actually be built—on a foundation of deep, specific domain data.1:21 This shift from raw power to applied purpose is forcing a much harder conversation.1:28 And that conversation just exploded between Anthropic and the Pentagon.1:33 CEO Dario Amodei is publicly resisting Pentagon demands for unrestricted use of Claude in military applications...1:42 including mass surveillance and autonomous weapons.1:46 The Pentagon’s response was to label Anthropic a "supply chain risk," and the Trump administration followed by halting federal use of their tech.1:57 Amodei is holding a line, insisting on guardrails.2:01 The Pentagon insists existing laws are enough.2:05 This is the conflict that defines the next era.2:08 It's not about if the AI can do something, but if it should.2:13 And we just got a field report on that exact question.2:17 A career diplomat, cited by Gary Marcus, reflected on the recent US-Iran conflict.2:23 The consensus?2:24 AI was useful for operations...2:27 but it was not good on strategy.2:29 Generative AI can process logistics and manage tactical support.2:34 It cannot, however, project into novel situations or build the kind of robust world model needed for strategic foresight.2:44 It lacks judgment.2:45 So as the industry pivots from bigger models to smarter agents, the highest-stakes customer is discovering the technology’s fundamental limits.2:56 The real story isn't about building a better brain; it's about deciding what kind of work we trust it to do.
