Why nerds may have to change in the age of AI: David Richards
Bill Gates was not joking when he said: “Be nice to nerds. Chances are you’ll end up working for one.” For three decades, code was leverage. If you could build software, you could build distribution.
If you could build distribution, you could build a business. And if you could build a business, you could become extraordinarily rich.
That is why I used to give young people the same advice in different words. Learn to code. It is the key. I even helped open a coding academy because I believed it so completely.
This week, that belief took a direct hit. Elon Musk has predicted that traditional coding will be obsolete by the end of this year. Hyperbolic perhaps, but not unserious. At the World Economic Forum in Davos last month, Dario Amodei, the chief executive of Anthropic, told The Economist that AI could handle “most, maybe all” of what software engineers currently do within six to twelve months. His own chief product officer confirmed days ago that at Anthropic itself, the code is now effectively 100 per cent AI-generated.
I agree with Musk. Not because every programmer vanishes by next Christmas, but because the labour market for programming is about to be structurally rewritten.
To understand why, you have to understand what coding really was in the Internet era. It was permission. Creative people had ideas, but they needed a translator who spoke machine. Designers, marketers, entrepreneurs and domain experts all hit the same wall. You can imagine it, but can you build it?
Take Facebook. The mythology of Mark Zuckerberg as sole genius is well-rehearsed. The reality is more instructive. Harvard students Cameron and Tyler Winklevoss and Divya Narendra had conceived a social network called HarvardConnection and needed a programmer to build it. They recruited Zuckerberg. He agreed to help. Then he launched his own version. The dispute ended in a $65m settlement.
Whatever view you take of that episode, one fact stands out. The creative spark was not the scarce resource. The ability to execute in code was. That asymmetry gave the nerd the power. That is the part that has now changed.
AI can already produce working software of startling quality. What once required a team of engineers working for weeks can increasingly be achieved by a single person with a well-crafted prompt and good judgment. The machine speaks fluent software. The translator is no longer the bottleneck.
Which means the old Gates line needs updating. For years, being the person who could write the code gave you leverage over the person who merely had the idea. Now the advantage is shifting back. The person who can define the problem clearly, exercise taste, and demand quality from increasingly capable machines may hold more power than the person who once held the keyboard.
Some programming will remain. Safety-critical systems. Deep infrastructure. Security architecture. But the mass-market demand for armies of developers writing routine business software is the part most exposed to replacement.
Be nice to nerds, Gates said. The advice still holds. It is just that the nerds of tomorrow may look nothing like the nerds of today.
David Richards MBE entrepreneur and philanthropist