0:00 OpenAI just raised 122 billion dollars, valuing the company at 852 billion as of April first.0:07 The checks are staggering: 50 billion from Amazon, 30 billion from Nvidia, and another 30 billion from SoftBank.0:17 OpenAI says the capital is for building "the infrastructure layer for intelligence itself," a foundation for the entire economy.0:28 This round anchors a record-breaking quarter.0:32 In just the first three months of 2026, AI companies collectively raised 297 billion dollars—shattering the previous year's pace.0:43 But while OpenAI chases scale with capital, others are chasing efficiency.0:49 Liquid AI just released LFM2.5-350M, a compact model with only 350 million parameters.0:56 The key is that it was trained on an enormous 28 trillion tokens, outperforming models twice its size.1:05 Meanwhile, Facebook's research division announced an AI agent called AIRA_2.1:11 It was tested on MLE-bench, a suite of real-world data science challenges, and surpassed human expert performance in a single day.1:22 Google also continues its push, with multiple updates from DeepMind and Google Research announced throughout March.1:32 So let's connect these threads.1:35 On one side, you have OpenAI operating at a scale that is difficult to comprehend.1:42 Nine hundred million weekly users.1:44 A two billion dollar monthly revenue rate.1:48 But the compute costs are just as astronomical.1:52 This 122 billion dollar raise isn't a victory lap.1:56 It’s a survival mechanism.1:58 Reuters reports that competitive pressure from Google and Anthropic has forced OpenAI to redraw its product roadmap twice, redirecting resources to revenue-generators like Codex and other business tools.2:16 The need for this funding is a signal of both their ambition and their vulnerability.2:23 Then you have two developments that directly undermine that entire strategy.2:29 First, Liquid AI's model.2:31 It doesn't use a pure Transformer architecture.2:35 It uses a hybrid backbone called LIV, designed for 'intelligence density.' It proves that a 350-million parameter model—small enough for edge devices—can beat larger models if the architecture is right and the training data is massive enough.2:56 This isn't just an iteration.2:58 It's an argument that smarter design can beat bigger budgets.3:04 Second, there's Facebook's AIRA2.3:06 The real breakthrough is its operational autonomy.3:11 Previous AI agents doing research would stall.3:14 They’d hit a hardware glitch or a bad experimental path and require human intervention.3:22 AIRA2 just… works around it.3:24 It runs experiments in parallel across multiple GPUs and scales its output linearly with more hardware.3:33 This shifts the bottleneck from human ingenuity to hardware availability.3:39 You're no longer just buying compute to run a model; you're buying compute to automate discovery itself.3:48 So we now have two distinct philosophies competing for the future of AI.3:54 One believes the path forward is paved with unprecedented amounts of capital, building ever-larger infrastructure.4:04 The other is quietly proving that the right architecture—and the right automation—can make that capital irrelevant.
