Key Takeaways:
Anthropic’s push to automate white-collar work is starting at home, with the company’s own AI models now handling 90% of its internal software development and financial reporting. The move signals a profound shift in how knowledge workers operate, turning employees from task-doers into supervisors of AI systems and accelerating a high-stakes race for enterprise clients against rivals like OpenAI.
"90 plus percent of our code is actually written by Claude," Anthropic CFO Krishna Rao said on a recent episode of the "Invest Like the Best" podcast. He described a workplace where AI handles the bulk of execution, freeing up human employees to focus on oversight, judgment, and strategy, which he argues increases the company's "talent density."
The transformation is stark. Rao noted that on the finance team, the monthly financial review process is "90 to 95% ready" before humans intervene to interpret the results. Tasks that previously took hours are now completed in as little as 30 minutes. This productivity surge is a core part of Anthropic’s strategy, allowing the company to get more done and making new hires more effective as they learn to work alongside Claude.
The trend has significant implications for investors and the broader tech landscape. As Anthropic demonstrates massive internal efficiency gains, it builds a powerful case for enterprise customers weighing AI adoption. The company, which has seen its market share among business users quadruple since May 2025, is reportedly raising a new funding round that could value it near $950 billion, potentially eclipsing OpenAI’s last valuation.
The shift inside Anthropic suggests the nature of white-collar work is changing rapidly. Rao’s comments paint a picture where employees increasingly oversee fleets of AI agents rather than completing tasks manually. "Everyone kind of becomes a manager," he said, arguing that as employees become more productive with AI assistance, companies may expand hiring because "there's no shortage of work to do."
This new management role, however, requires a deep level of expertise. Cat Wu, Anthropic’s head of product for Claude Code, emphasized that managing AI agents effectively requires the ability to do the job oneself. "It is extremely hard to manage agents if you can't do the job yourself," Wu said in a recent interview. "You have to have the ability to debug it... Was my request under-specified?"
Anthropic's accelerating adoption is driving an urgent need for more computing power. The company recently signed a $1.8 billion, seven-year cloud deal with Akamai Technologies Inc. to support its expanding AI infrastructure. This follows earlier agreements with Google and SpaceX to meet the intense demand for its Claude models, which have gained significant traction in coding and enterprise automation.
The company's rapid pace of development is set to continue. "The models are still improving at a very steady pace," Wu said, hinting at further capabilities. The next frontier, she suggests, is proactivity, where Claude will not just respond to requests but will anticipate user needs and set up automations independently. This vision for the future, combined with a relentless focus on safety, as seen with the controlled release of its powerful cybersecurity model Mythos, positions Anthropic as a formidable force in the AI industry.
This article is for informational purposes only and does not constitute investment advice.
