Anthropic's Cat Wu Predicts Shift to Anticipatory AI and Agent Management
At the second annual Code with Claude conference, Wu detailed a strategy focused on the technological frontier over competitive benchmarking.
Primary source: TechCrunch AI. Full source links and update notes are below.
Fast summary
Start here
- Anthropic is reportedly seeking a valuation of $950 billion as it gains significant market share among business customers.
- Product lead Cat Wu describes the future of work as managing 'fleets of agents' where human domain expertise is required for debugging.
- The company is maintaining a restricted release strategy for powerful models like the Mythos cybersecurity tool to prevent weaponization.
What happened
Cat Wu, Anthropic’s head of product for Claude Code and Cowork, outlined the company's long-term product vision during the second annual Code with Claude conference in San Francisco. Wu detailed how the company approaches development by focusing on the 'exponential' growth of AI capabilities rather than reacting to the release cycles of competitors like OpenAI.
What's new in this update
Wu clarified Anthropic's stance on the burgeoning agent economy, suggesting that the future of professional work involves humans managing vast fleets of AI agents. She emphasized that this transition requires a new skill set akin to traditional people management, where experts must be able to 'debug' agent decisions and refine underspecified requests to ensure accuracy.
Key details
Anthropic is currently looking to raise tens of billions of dollars at a valuation that could reach $950 billion, potentially surpassing OpenAI's recent $854 billion valuation. This financial trajectory is supported by a surge in enterprise adoption; recent reports indicate Anthropic has quadrupled its market share among business customers since May 2025, with many users increasingly preferring Claude over ChatGPT for technical and coding tasks.
Background and context
Since joining Anthropic in August 2024, Wu has been instrumental in evolving Claude from a standard chatbot into a suite of specialized coding and collaboration tools. This strategic focus includes the Glasswing initiative, which launched in April. Glasswing provides a template for the restricted deployment of high-risk models, such as the Mythos cybersecurity model, which is available only to a small consortium of partners like Amazon, Apple, and CrowdStrike due to concerns about its potential use by bad actors.
What to watch next
Anthropic expects to continue its rapid pace of model development, having already matched its 2025 output in the first half of 2026. Future deployments are likely to follow the tiered access model seen with Glasswing, where the most powerful tools are kept from the general public until safety guardrails are deemed sufficient.
Why it matters
As Anthropic nears a trillion-dollar valuation, its product philosophy on agent autonomy and restricted safety releases sets a precedent for the enterprise AI industry.
Read next
Follow this story through the topic hub, more ai coverage, and the latest updates.
Weekly briefing
Get the week's key developments in one concise email.
Get a fast catch-up on the biggest stories, the context behind them, and the links worth your time.
Cadence
Weekly, for a quick catch-up
Coverage
AI, business, world, security, sports
Format
Clear takeaways and useful context
Request the briefing
Leave your email to open a prepared request and get on the list for the weekly briefing.
Author
See who assembled this story and follow more of their work.
Sources and methodology