Anthropic eyes massive valuation as it gains on OpenAI

Anthropic may raise tens of billions of dollars in a funding round that would value the company at some $950 billion. This valuation would place the San Francisco-based artificial intelligence company ahead of its competitor OpenAI, which was valued at $854 billion in March. The potential funding comes as Anthropic gains ground on its rival, having quadrupled its market share among business customers since May 2025 to outpace OpenAI in that segment.

To maintain this momentum, Anthropic is keeping up a rapid product release cycle. The company released at least six models last year and has released almost as many this year. Cat Wu, Anthropic’s head of product for Claude Code and Cowork, explained that the team’s product strategy is focused on keeping pace with rapid technological improvements. “The main thing that we design for is staying on the exponential, so I think, across our team, we instill in everyone the lesson that AI will just continue to get better,” Wu said. Over the next six months, the company plans to focus on proactivity—an AI capability to anticipate user needs where the model, Claude, understands what users are working on and automatically sets up automations for them.

While Anthropic aims to deploy its models widely, some releases require stricter controls. In April, the company launched Glasswing, a cybersecurity initiative that grants restricted access to its new cybersecurity model, Mythos, to a small consortium of partner organizations, including Amazon, Apple, CrowdStrike, and Microsoft. Unlike its other models, Mythos is not being given a general public release. Anthropic has withheld a public release because it fears the model is too powerful and could be weaponized by bad actors.

This controlled rollout aligns with Anthropic’s broader vision for the workplace, which Wu describes as a shift toward human staff managing fleets of agents—AI systems capable of performing tasks or managing workflows. However, Wu cautions that this transition will require human supervisors to maintain deep technical expertise. She noted that it is extremely difficult to manage agents if a person cannot perform the job themselves, emphasizing that managers must remain domain experts to debug and oversee these automated systems. Wu frequently collaborates on these product developments with Boris Cherny, a core member of Anthropic’s technical staff and the creator of Claude Code, a partnership that has led the duo to be characterized within the company as Batman and Robin.

Why it matters

Anthropic is scaling its operations and product strategy, aiming to potentially surpass OpenAI in valuation while shifting its focus toward proactive AI agents that automate user workflows.