In a technology landscape increasingly dominated by artificial intelligence models, Anthropic is experiencing a period of remarkable growth and success.
The company is poised to potentially surpass its primary competitor, with plans to secure tens of billions of dollars in an upcoming funding round that could push its valuation to approximately $950 billion. This figure would exceed OpenAI's valuation of $854 billion from its March round. Furthermore, business clients are increasingly favoring Anthropic's Claude over ChatGPT; a recent report indicated that Anthropic has outpaced OpenAI among business customers, quadrupling its market share since May 2025.
A pivotal figure in this ascent is Cat Wu, Anthropic’s head of product for Claude Code and Cowork. Since joining the company in August 2024, Wu has been instrumental in guiding Claude through a crucial developmental phase, transforming it from a purely informational chatbot into a sophisticated coding tool and beyond. Wu, who oversees new feature development, frequently collaborates with Boris Cherny, a core member of Anthropic’s technical staff and the creator of Claude Code, a partnership that has earned them the moniker Anthropic’s “Batman and Robin.”
Wu recently participated in an interview at the second annual Code with Claude conference in San Francisco, where she delved into her product strategy philosophy and her vision for the future evolution of the Claude user experience.
This interview has been edited for length and clarity.
When questioned about the extent to which product strategy is reactive to competitors, Wu asserted that Anthropic’s primary design principle is to “stay on the exponential.” She explained that the team is instilled with the understanding that AI will continuously improve, necessitating a constant presence at the frontier of innovation. Wu emphasized that Anthropic intentionally avoids focusing on competitors, stating that such a focus can lead to being “perpetually two weeks, or like, a month behind how fast you can execute.” She believes this approach is crucial for remaining at the cutting edge of AI development.
Regarding Anthropic’s rapid development pace, with at least six models released last year and nearly as many already this year, Wu expressed hope that this tempo would continue (with a laugh). She noted that models are still improving steadily, allowing the company to keep sharing these advancements with users. While deployments might evolve, as seen with initiatives like Glasswing, the overarching goal is to safely extend the benefits of this intelligence to as many people as possible.
Glasswing, an initiative launched by Anthropic in April, involved inviting a select consortium of partner organizations—including major companies such as Amazon, Apple, CrowdStrike, and Microsoft—to access its new cybersecurity model, Mythos. Unlike many of Anthropic’s other AI models, Mythos has not been made generally available to the public. The company cited concerns that the model, designed to scan codebases for software vulnerabilities, is exceptionally powerful and could potentially be weaponized by malicious actors.
Addressing a previous statement from Wu that the future of work involves staff managing “fleets of agents,” the interviewer posed a concern that agents might eventually become more proficient than humans at their jobs. Wu countered that it is “extremely hard to manage agents if you can't do the job yourself,” stressing that managers must remain experts in their respective domains. She characterized managing agents as a new, essential skill set, akin to managing people, requiring the ability to debug and understand agent errors, such as misinterpreting instructions or under-specified requests.
When asked if the long-term goal of agent deployment was to reduce team sizes, potentially eliminating roles like interns, Wu offered a different perspective. Ideally, she believes the aim is to empower everyone to achieve significantly more. She highlighted that every job contains a percentage of “really tedious” tasks—for her, it’s responding to emails. Wu's hope is that AI agents will handle these mundane aspects, freeing up individuals to pursue more creative and impactful endeavors in their newfound spare time.
Looking ahead to the next six months, Wu expressed particular excitement about the concept of “proactivity.” She noted a shift from last year’s synchronous development to current routines, such as automating customer support ticket responses. The next evolutionary step, she envisions, is for Claude to understand users’ work patterns and autonomously set up relevant automations for them.
The Editorial Staff at AIChief is a team of professional content writers with extensive experience in AI and marketing. Founded in 2025, AIChief has quickly grown into the largest free AI resource hub in the industry.