IPO Revealed: Anthropic's Next Step

Summary of IPO Revealed: Anthropic's Next Step

by Candace Fan

13m•June 1, 2026
Listen to Episode

Overview of IPO Revealed: Anthropic's Next Step

This episode covers a fast-moving set of AI and tech developments, led by Anthropic’s confidential IPO filing and broader signs that the AI industry is entering a new phase of competition, monetization, and infrastructure scaling. The discussion also touches on Microsoft building more proprietary AI capabilities, NVIDIA’s push into robotics reasoning, Intel’s renewed AI chip strategy, Strava monetizing API access, and a weather startup using proprietary balloon data to outperform traditional forecasting systems.

Anthropic’s Confidential IPO Filing

  • Anthropic has reportedly filed confidentially for an IPO on June 1.
  • The company is said to have just completed a $65 billion Series H at a $965 billion valuation as stated in the transcript.
  • If accurate, that would make Anthropic one of the most valuable private companies in the world and ahead of OpenAI’s reported valuation.
  • The episode frames this as a major milestone in the AI industry’s “battle royale” for dominance among Anthropic, OpenAI, and SpaceX.
  • Anthropic’s revenue is described as surging rapidly:
    • $9 billion run rate at the end of last year
    • $47 billion run rate now, according to the transcript
  • The confidential filing lets Anthropic refine financials and risk disclosures privately before public release, which is standard for large IPO candidates.

Strategic context

  • Anthropic is also said to pay over $15 billion annually for data center capacity, highlighting how dependent frontier AI companies remain on infrastructure partners.
  • The transcript suggests this IPO could set the tone for how other mega-cap private companies approach public markets.

Microsoft’s MAI Thinking One and AI Independence

  • Microsoft unveiled MAI Thinking One, described as its first reasoning model built without relying on OpenAI or Anthropic.
  • The launch was presented as a signal that Microsoft wants to reduce dependency on external model providers, especially in enterprise settings.
  • Microsoft emphasized that the model was not trained via model distillation from another company’s outputs.
  • The company also introduced:
    • A Windows 11 developer mode
    • Pre-installed tools and a distraction-free environment
    • New image models, including MAI Image 2.5 and MAI Image 2.5 Flash
  • Overall, the message is that Microsoft wants to be seen as a stronger standalone AI platform, not just a reseller of OpenAI capabilities.

NVIDIA’s Cosmos 3 and the Robotics Race

  • NVIDIA introduced Cosmos 3, a foundational model for physical AI reasoning.
  • Its purpose is to help robots reason about actions before executing them in real-world settings.
  • The model works by:
    • Simulating possible outcomes using a learned world model
    • Selecting the highest-probability action
    • Improving autonomous decision-making in robotics
  • NVIDIA positioned Cosmos 3 alongside its Isaac and Omniverse tooling, reinforcing its integrated robotics stack.
  • The episode frames this as part of a broader race with:
    • Google DeepMind
    • Physical Intelligence
    • Other embodied AI companies
  • Main takeaway: the next major AI frontier may be robotics and embodied intelligence, not just chat or cloud AI.

Intel’s Crescent Island Chip and the Comeback Narrative

  • Intel announced Crescent Island, an AI chip expected to ship later this year in limited quantities.
  • Intel is targeting inference workloads, where price sensitivity matters more than raw training performance.
  • Key differentiators:
    • Uses cheaper LPDDR5 memory
    • Uses air cooling instead of expensive HBM and liquid cooling setups
  • The chip is positioned as a lower-cost alternative to NVIDIA and AMD offerings.
  • Intel is also developing the chip in its own fabs, reinforcing a vertically integrated strategy.
  • The transcript notes that:
    • Intel’s stock has surged more than 200% this year
    • New CEO Lip-Bu Tan is aggressively rebuilding the AI silicon roadmap
  • Another potential advantage: the chip may be exportable to China under current U.S. export controls, opening a revenue path that is harder for NVIDIA and AMD to access.

Strava’s API Paywall and the Monetization of Data Access

  • Strava is putting its API behind a $11.99/month paywall.
  • The company says the change is driven by a 448% spike in developer applications and the rise of zero-code AI app builders.
  • Their argument: free API access is being overused by third parties that consume infrastructure without contributing to costs.
  • Important distinction:
    • Wearable integrations like Garmin, Apple Watch, and Wahoo remain free
    • Third-party app builders now have to pay
  • Strava also launched a direct integration with Anthropic’s Claude, allowing users to ask AI questions about workout data without paying separately.
  • The episode compares this move to similar API monetization shifts from Reddit and X/Twitter.

Weather AI: WindBorne’s WeatherMesh 6

  • WindBorne Systems released WeatherMesh 6, an AI weather model that reportedly outperforms the ECMWF forecasting system, often considered the gold standard.
  • Its key advantage is proprietary data collection:
    • 400 balloons feed raw atmospheric sensor data directly into the model
    • This bypasses traditional government data pipelines
  • The model offers:
    • Six updates per hour
    • 3 km resolution
    • Coverage across Europe and the continental U.S.
  • WindBorne’s balloon network operates from 15 global launch sites.
  • Customers mentioned include:
    • NOAA
    • The U.S. Air Force
    • The U.S. Navy
    • Commodity traders
  • The takeaway: proprietary physical data collection is becoming a major AI moat, especially in domains like forecasting.

Key Takeaways

  • Anthropic’s IPO filing signals that frontier AI companies are moving from growth stories to public-market-scale businesses.
  • Microsoft is trying to become a more independent AI platform by building reasoning and image models in-house.
  • NVIDIA is pushing deeper into robotics, where embodied reasoning may become the next major AI battleground.
  • Intel is betting on lower-cost inference chips and export flexibility to regain relevance in AI hardware.
  • Strava is showing that API access is increasingly treated as a paid product, not a free utility.
  • WindBorne demonstrates how proprietary real-world data can outperform legacy forecasting systems.

Overall Theme

The episode’s broader message is that the AI market is entering a new stage: not just model building, but IPO readiness, infrastructure economics, vertical integration, data monetization, and real-world deployment.