Why College Degrees Don’t Matter Anymore

There has always been the meme of college dropouts running Silicon Valley companies. But for every Mark Zuckerberg, Silicon Valley is filled with dozens of founders like Jeff Bezos or Larry Page or Sergey Brin — highly educated founders with a bachelor’s degree or more.

This is changing. College dropouts are all the rage, as they are running some of the most successful AI companies of this moment: Companies like Mercor, Etched, Scale (now MSL), and more. It’s not just that these companies are run by college dropouts, but they are being staffed by a combination of dropouts and inexperienced hires who join right out of school. Even OpenAI, Anthropic and other big labs are increasingly hiring dropouts at the junior level, not to mention Cursor, whose talent pipeline remains a core competitive advantage. This is a big change from five or ten years ago, when Google and Facebook alumni were staffing many of the prominent, up-and-coming companies.

Why now?

As always with these types of changes, there is a pull and a push.

The pull is the gravitational force that is the AI revolution. Young people feel like they are living through a once in a generation or even once in human history moment, and they want to be a part of it. Equally important, they feel that they are capable of contributing. Today’s AI startups don’t require the level of technical expertise that the last decade required (because so much of the heavy lifting is done at the foundation model layer). What they require is velocity, innovation, nimbleness — something young people have in spades. Even the big labs themselves are increasingly relying on on-the-job training, because the main experience that’s valuable to them is reps doing big GPU runs.

The push is that college is not as useful as it used to be. In the 2010s, studying computer science made sense, because knowing how to code was an extremely useful skill, and one that had to be trained. Today, AI does most of the coding anyways and junior engineers are not considered to be very valuable — so you might as well be a jack-of-all-trades who learns on the job. Meanwhile, cheating has become so rampant on college campuses that it begs the question of what the whole point of the experience is all about. Learning how to think still matters, but it can be done outside of the classroom.

It’s gotten so extreme now, that increasingly many founders and venture capitalists are weighing high school credentials more highly than college credentials in their evaluation of founders and potential employees. These are credentials like winning the International Math Olympiad or winning the National Debate Championship or excelling athletically and playing soccer at an international level. Why is this? These types of gauntlets demonstrate extreme ambition, competitiveness, and importantly, they cannot be faked. College used to represent a similar gauntlet, but the evidence suggests many people don’t believe this to be true anymore. You hear stories all the time right now in the Valley of Ivy League graduates who can’t do basic probability math in job interviews or computer science graduates who can’t solve entry-level coding problems. Whether true or exaggerated, this has undermined the value of the credential.

There are those who will push back, and they will say that college is a social experience and that depriving young people of this experience is bad. There’s truth to that argument, but I think it’s up to the students to decide what they want. Additionally, there is a new social scene in SF — perhaps one that was always there but is now greatly expanded — where founders are supporting each other, living together, and creating another social ecosystem, with its own benefits to human growth. Others will say that learning still matters and college teaches you how to learn — I strongly agree with this, and my critique is not of education itself, but rather with the pace at which today’s institutions will be able to adapt.

A final source of pushback will come from parents who ask, “What will happen when the AI bubble bursts?” Even if this is 1999 and we’re in the middle of the dot-com bubble, in hindsight, I think it was still the right move to get over to Silicon Valley and get into the action. Even more so if AI is indeed the defining technology of our lifetimes, as many people think. It is a testament to the power of being in the right place at the right time that so many of the leading figures of the dot-com era remain the leaders of Silicon Valley today — people like Elon Musk and Peter Thiel and Marc Andreessen. Being in the action during a paradigm shift gets you a seat at the table; it’s much harder to earn that seat during the doldrum years.

In every age, young people think and act for themselves, and they challenge existing norms. In a time when the world is moving at lightning speed, it’s unsurprising that young people see something obvious: Big institutions cannot change at the pace of AI. The old paths to stability — become a lawyer or an engineer or a consultant — are no longer a sure thing. No wonder that many are dropping out to be part of the AI revolution, believing that if they are part of the change, they can find a stable place for themselves in the new equilibrium.

I don’t think dropping out is the only path. In fact, one hope I have is for a new type of education system to emerge. It’s not that higher education is inherently flawed, but that our institutions cannot adapt at the pace required to prepare young people for a future that is evolving so quickly. Education is not dead, but it needs to be reinvented. A new education system would need to focus on things that humans are uniquely good at — creating new knowledge, not regurgitating old knowledge, understanding our fellow humans, not staring at computers, contemplating transcendence, not nihilism. Maybe this new system is something that the next batch of college dropouts will figure out for us.