Dario Amodei says that his prediction of AI writing 90% code by this month has come true at his company — but it hasn’t led to the job losses that many had anticipated.
The Anthropic CEO recently shared insights about how artificial intelligence has transformed software development within his own organization. Speaking about the rapid adoption of AI-generated code, Amodei revealed that the shift has been both more dramatic and more ordinary than outside observers might expect, challenging common assumptions about AI’s impact on tech employment.
“I would say maybe 70, 80, 90% of the code written at Anthropic is written by Claude,” Amodei stated. He acknowledged that when he made similar predictions “three or six months ago, people think of it as falsified because they think of it as we’re gonna fire 70, 80, or 90% of the software engineers.”
However, the reality at Anthropic tells a different story. “But what really happens is that humans become managers of AI systems,” Amodei explained. “There’s a shift because of the principle of comparative advantage. So it looks more normal than you think.”
The CEO reflected on the broader pattern of technological transformation, noting that dramatic changes often feel less revolutionary than anticipated once they arrive. “I think eventually that logic may not hold, but there’s a sort of sci-fi sheen to these predictions, to looking at the future. That it’s gonna be weird that you’ll be looking through different colored glasses, that there will be that’ll look like Star Wars or something,” he said. “Often when these predictions come true, it’s wild, but it’s also in a way ordinary.”
Amodei’s observations align with a growing trend across the tech industry, where AI coding assistants like Claude, OpenAI’s Codex and platforms like Cursor and Windsurf have become increasingly sophisticated. Recent surveys suggest that over 70% of developers now use AI tools regularly, with productivity gains reported across major tech companies. Rather than replacing programmers entirely, these tools appear to be reshaping the role into something more akin to AI supervision and high-level system architecture. The principle of comparative advantage that Amodei references suggests that humans will continue to focus on tasks where they maintain an edge over AI – strategic thinking, complex problem-solving, and managing AI systems themselves – while AI handles more routine coding tasks. This transformation represents not just a technological shift, but a fundamental reimagining of what it means to be a software engineer in the age of artificial intelligence.