As artificial intelligence continues to reshape industries and career paths, Anthropic cofounder Jack Clark is pushing back against the increasingly popular notion that only technical degrees hold value in the age of machine intelligence.

Speaking at Semafor’s World Economy Summit, Clark made the case that liberal arts disciplines, often dismissed in conversations about the future of work, may in fact be uniquely suited to the demands of the AI era. His argument goes beyond a defense of humanities education; it speaks to a broader reordering of what employers in frontier technology companies now prize most: judgment, synthesis, and the ability to frame consequential questions.

Clark’s own professional trajectory gives weight to that argument. Before helping build one of the world’s leading AI companies, he worked as a journalist and studied literature at the University of East Anglia, a background that might once have seemed far removed from advanced machine learning.

“What turned out to be useful is that I got to learn a lot about history and a lot about the kind of stories that we tell ourselves about the future,” Clark said on Monday during Semafor’s World Economy Summit. “That’s turned out to be like, extremely relevant for AI in a way that I think people wouldn’t have predicted.”

That remark goes to the heart of an increasingly important debate within the technology sector: as AI systems become more capable, the premium is shifting from narrow technical execution to contextual intelligence. History, philosophy, literature, journalism, and political economy train people to interrogate assumptions, understand human behavior, and interpret narratives about risk, progress, and power. Those are precisely the issues now confronting AI companies as they navigate regulation, ethics, safety, and societal disruption.

Rather than elevating any single discipline, Clark argued that the most valuable academic pathways are those built around intellectual overlap.

“I think that majors which are going to become more important are ones which involve like synthesis across a whole variety of subjects and analytical thinking about that,” he said.

This emphasis on synthesis is notable, especially at a time when AI is moving from a purely engineering challenge into a multidisciplinary enterprise involving law, public policy, security, philosophy, linguistics, economics, and behavioral science. Companies at the frontier are no longer merely building models; they are designing systems that interact with society at scale.

Clark went further, identifying the most valuable skill not as coding itself, but as intellectual discernment.

“The really important thing is knowing the right questions to ask and having intuitions about what would be interesting, colliders, different insights from many different disciplines,” he said.

That insight carries particular resonance in today’s labor market. As generative AI automates an increasing share of repetitive technical work, the advantage is moving toward those who can define the problem, challenge assumptions, and connect disparate streams of knowledge into a coherent framework. In effect, the ability to ask a better question may now be more economically valuable than the ability to execute a routine answer.

Clark’s comments on programming underscore that shift. After repeated pressing, he said that “rote programming” is something he would avoid, a statement that aligns with growing expectations that AI-assisted development tools will absorb much of the repetitive coding workload.

“Some people need to know those fundamentals, but we do see that technology move up the stack,” Clark said.

He is not dismissing technical foundations outright; rather, he is signaling that the nature of technical work is changing. Low-level, repetitive coding tasks are increasingly being automated, while higher-order functions such as architecture, systems design, product reasoning, and ethical oversight are becoming more central.

Perhaps the most striking line from Clark’s remarks was his reference to philosophy graduates working inside Anthropic, a statement that directly challenges long-standing assumptions about employability in the humanities.

Overall, he said, majors that may appear disconnected from AI are likely to remain highly relevant, noting that Anthropic employs philosophers.

“When was the last time you heard that a philosophy degree was like a great job prospect?” he said.

The deeper significance of that comment lies in what it reveals about where AI is heading. As models become more powerful and their social consequences widen, companies need people who can think rigorously about reasoning, ethics, alignment, human values, and unintended consequences. Those are questions philosophy departments have wrestled with for centuries.

Clark’s remarks, therefore, amount to more than career advice for students. They denote a structural shift in the AI economy, one in which interdisciplinary reasoning and humanistic inquiry are becoming assets rather than peripheral skills.

For students weighing their academic choices, the message is that the future of AI may belong not only to engineers, but also to those trained to understand people, ideas, institutions, and the stories societies tell about what comes next.