Top Anthropic Engineer Warns AI Agents Will Disrupt Computer Jobs, Calling Change 'Painful' and Inevitable
A senior engineer at Anthropic has issued a stark warning about the rapid transformation of computer-based jobs, saying that the emergence of advanced AI agents capable of performing tasks across digital platforms will soon disrupt nearly every internet-driven profession in the United States. Boris Cherny, the creator of Claude Code at Anthropic, the company behind the Claude AI chatbot, made the remarks during a recent appearance on Lenny Rachitsky’s podcast. Cherny emphasized that AI systems capable of autonomously operating computer tools—such as writing code, managing workflows, analyzing documents, sending messages, and building websites—are advancing at an unprecedented pace. These AI agents, he said, are moving beyond simple text generation to actively interacting with software environments in ways that mimic human workers. “It's going to expand to pretty much any kind of work that you can do on a computer,” Cherny said. “In the meantime, it's going to be very disruptive. It's going to be painful for a lot of people.” Claude Code, Anthropic’s AI coding agent built on its Claude models, is one of the most visible examples of this shift. The tool, which recently received updates with the release of Opus 4.6 in early February, enables AI to execute complex programming tasks, automate workflows, and even generate entire websites—functions that once required skilled human developers. Unlike traditional chatbots that only respond to queries, AI agents like Claude Code can take initiative, make decisions, and complete multi-step processes across multiple applications. While Anthropic acknowledges that current models still fall short of matching the precision and judgment of experienced human professionals, Cherny noted that progress is accelerating rapidly. He shared that his own team now relies heavily on AI to boost productivity, with engineers achieving significantly more work in less time since the rollout of Claude Code. He predicted that the capabilities of these systems will continue to improve, potentially rendering many current roles obsolete. In a separate interview with Y Combinator’s “Lightcone” podcast, Cherny went further, suggesting that the job title “software engineer” could begin to disappear by 2026 as AI takes over routine and even complex development tasks. The societal implications of this shift remain unclear, he cautioned. “As a society, this is a conversation we have to figure out together,” Cherny told Rachitsky. “Anyone can just build software anytime.” For workers facing this transformation, his advice was straightforward: engage with AI tools early, understand how they work, and adapt. “Don’t be scared of them,” he said. “Learn how they can help you.”
