0:00 Meta just launched two new models, Muse Wonder and Muse Wonder Full.0:04 In our last episode, we talked about finding the signal in the noise.0:09 This is all signal.0:11 Not because of the performance scores, but because Meta just abandoned its open-weight strategy.0:18 That's the real story, and it changes the competitive landscape.0:22 But first, the rest of the field.0:25 US AI data center startups have now raised thirty-three point five billion dollars this year.0:31 That’s nearly double the total for all of 2025.0:35 But the money isn't just for more chips.0:38 A new Morgan Stanley report identifies the real bottleneck: Power.0:42 Over eighty percent of that funding is going toward expanding power grids and improving computing efficiency per watt.0:51 The gold rush has moved from the processors to the power plants.0:56 Meanwhile, Toptal just launched a platform called ExpertCrowd.1:00 The entire premise is to train and evaluate AI models with vetted human domain experts.1:06 They also released a new software engineering benchmark that exposes significant deficiencies in leading models like Opus five point five and Fable five point one.1:18 It is a stark reminder that expertise, not just brute-force scale, is required to achieve precision on complex tasks.1:27 At the enterprise level, the HR Tech 2026 conference saw a wave of new AI products.1:33 But this wasn't a revolution.1:35 The focus was on governance, compliance, and AI-driven screening with tamper-evident decision records.1:42 This is maturation, not a breakthrough.1:45 It’s about making AI usable and defensible for business, not changing the paradigm.1:51 And finally, a reality check from the world of physical AI.1:55 Robotics experts are pushing back timelines for artificial general intelligence milestones to 2030 or even 2031.2:03 The reason?2:04 Dexterity.2:05 The CEO of RLWRLD put it bluntly: "hardware and software for robot brains and hands must co-evolve." Building machines that can interact with the physical world is proving much harder than just scaling up language models.2:21 Okay, let's go back to Meta.2:23 This is the move that defines the week.2:26 The new Muse Wonder Full model scores a 56 on the Artificial Analysis Intelligence Index.2:32 That puts it in a statistical dead heat with Anthropic's Claude Sonnet five point five, and comfortably ahead of OpenAI's GPT-6 Astra at 53.2:42 On paper, Meta is back at the frontier.2:45 But here’s why that matters less than the strategy.2:49 For years, Meta’s play was Llama.2:51 Open weights.2:52 Let the community build, let the ecosystem flourish around their architecture.2:58 With Muse, that's over.2:59 These models are closed.3:01 You can't download them.3:03 You can't self-host them.3:04 This is a direct pivot to compete with OpenAI and Anthropic on their own terms: as a closed, proprietary, frontier system.3:13 Why the change?3:14 Because the game at the absolute frontier is no longer about chatbots.3:19 It’s about building autonomous agents.3:22 Meta is explicit about this, highlighting features like "Active Memory" and "Active Focus." Active Memory lets the model retain relevant information across long, complex interactions.3:35 Active Focus helps the model navigate obstacles, recover from errors, and ask for clarification.3:42 These aren't features for a simple Q and A.3:45 They are foundational blocks for agents that can actually do things in the world.3:51 And to monetize that kind of capability, you need control.3:55 The open-source playbook just doesn't work for that business model.4:00 That ambition for more capable agents runs headfirst into a physical wall.4:06 And that brings us to the thirty-three point five billion dollars flowing into data centers.4:12 The Morgan Stanley report puts it plainly: "Power is the key constraint." This is the single most important sentence in AI infrastructure right now.4:23 Since 2024, over fifty billion dollars has been raised by startups in this space, focused on four key problems.4:31 First, simply expanding the power supply.4:34 Second, radically improving computing efficiency per watt.4:38 Third, integrating the power and computing systems so they work together, not as separate components.4:45 And fourth, completely rethinking data center design and location to get closer to power sources.4:52 This isn't just an engineering problem.4:55 It is an economic one.4:57 It means the cost to train and run a frontier model is no longer just about buying more GPUs.5:04 It's about securing multi-year power agreements and building entirely new infrastructure.5:10 This raises the barrier to entry enormously and explains why the field is consolidating around a few giant, capital-intensive players.5:20 It also explains why Meta, despite its history, decided it needs to play the same closed-system game.5:27 The era of easy scaling is over.5:29 The era of wrestling with physical limits has begun.5:33 So, what is the delta today?5:35 For the last few years, the story of AI has been about scale.5:40 More data, more compute, bigger models.5:42 That was the primary axis of competition.5:45 That era is now becoming table stakes.5:48 The new game is about physical constraints and specialized intelligence.5:53 It's a battle over watts, not just weights.5:56 It’s a race for human expertise to solve problems that raw data can't.6:01 The frontier is no longer just a digital space.6:05 It’s being defined by concrete, cooling pipes, and the global electricity grid.
