The Next Billion Developers
Artificial general intelligence (AGI) is Silicon Valley’s latest obsession. Brilliant minds like Elon Musk and Sam Altman believe achieving AGI is not just likely, but inevitable. Others disagree. Yann LeCun, one of the godfathers of machine learning, has said “there is no such thing as AGI.” Jack Clark of Anthropic has described the “breathless enthusiasm” for AGI as a “misplaced religious impulse.”
As a lightning rod, AGI captures our imagination. AGI promises free education and housing, the eradication of poverty with universal basic income, and a reallocation of human capital to science and art. These are worthwhile objectives.
But what if AGI is a distraction? What if there’s another path to this same future—one that does not require a technological leap of faith and is already underway?
There are more developers today than ever before. And the population is starting to snowball. Our thesis is that bringing 1B developers into the software economy could rival AGI in terms of its economic impact.1
Software engineering is one of the most powerful tools that humans have ever invented. If we can make it even more widely available—assisted and accelerated by AI—it could result in the automation of many non-tech industries, creating an economic surplus.
But is this target of 1B developers really achievable? Isn’t software engineering too hard to teach to so many people? And what would we do with all those developers?
Let’s start with the goal of 1B developers. This year, the developer population crossed 100M, up from 24M when Github first started reporting the metric in 2017.2 In Github’s recently published State of the Octoverse report, they shared that the global developer population grew 26% in the last year. Github added 21M new accounts in 2022.
We have three reasons to believe hypergrowth in the developer population will persist.
First, it’s been getting much easier to become a developer, and we expect this flywheel to continue. In just the last decade, Cloud abstracted away the server. Databases like MongoDB and Postgres simplified the back-end. Vercel and React reshaped the front-end. The best software engineers continue to focus on improving the discipline itself. By iteratively building better and better abstractions, software engineers have created a flywheel that reduces complexity while increasing potency. As a result, a 12 year-old in India can now build a web app in a matter of days, something that would have taken a professional developer weeks before.
Second, market forces continue to attract more and more young people to become software engineers. Already, the best and brightest are disproportionately learning to code. Rather than dampening as more talent has entered the field, the economic returns to software engineering only seem to be expanding. What this means is that the productivity of software engineers, and the applicability of their skills, is expanding faster than the supply.
Third, AI has been an unanticipated accelerator. Today, Github Copilot is writing 46% of users’ code. Zapier enables developers to build automations with text descriptions. Open Interpreter is turning the command line into a chat interface. With AI copilots, we may eventually be able to program computers using natural language alone.
Growth in the developer population enables the software economy to broaden its scope and impact. For example, companies like Stripe, Twilio and Plaid have created essential building blocks for payments, telecommunications and banking. Traditional companies in mature industries are bringing on more and more software engineers to drive internal digitization. Newer startups are also taking software and AI and applying these technologies to sectors like mining, robotics, construction, accounting and legal. The opportunity with 1B developers is to drive automation not just in the digital economy, but in the physical economy as well.
Unlike AGI, the promise of 1B developers is not about eliminating work altogether. Rather, it is about taking our resources—human resources—and leveraging them to have more and more impact. Thus, the story of the coming decades may be about the increased power of AI. But more quietly, the methodical march of the developer revolution will continue.
If you are building in this space, we’d love to hear from you. Please reach out at firstname.lastname@example.org.
- Amjad Masad was one of the first visionaries to articulate this goal and has been at its forefront
- Github’s definition: “developers are individual, not-spammy user accounts on GitHub”