by Wensupu Yang
"If you’re a reader, you are — secretly — a writer," the best-selling novelist Lee Child once said. To paraphrase: If you're a software user, you might also — secretly — be a program developer. Enabled by recent advances in artificial intelligence, specifically Large Language Models (LLMs), I want to explore a future where everyone programs.
The goal isn't to predict this future but to present signals and trends — some I've observed, some I’ve experienced firsthand — that show such a scenario is not only plausible but might be desirable. Along the way, I'll draw out some implications that could redefine our relationship with technology.
From Frustration to Creation
When I was 12, I begged my parents to buy me a "programmable" robot car. Unbeknownst to my younger self, the robot was programmed in C++, a low-level programming language challenging even for adult beginners. At the time, I barely spoke any English, and living in a small town, I had no one nearby to turn to for help. Frustrated, I gave up. Despite considering myself tech-savvy, I resigned myself to the notion that programming was reserved for others.
Fast forward to today, and the landscape has dramatically changed. Just last month, with the help of an array of AI assistants, I created five programs — some simple scripts with graphical interfaces, others fully fleshed-out web applications. What they all have in common is the speed of their creation: often, I have an idea in the morning and a working prototype by evening. For instance, after reading numerous biographies, I wanted a way to visualize the lifespans of influential figures on a timeline. Last Friday afternoon, I had this idea; by Saturday, I had a prototype up and running.
The State of Software Today: Winner-Takes-All
Software has evolved into a winner-takes-all market. Tech giants dominate due to near-zero marginal costs and the Matthew Effect — the rich get richer. This concentration leads to increased complexity, making it difficult for average users to tinker or develop their own solutions.
But this model has its drawbacks. Privacy concerns are rampant, incentives are misaligned, and we're often left with bloated software that offers features we don't need while lacking the one or two we truly desire.
AI as the Great Enabler
Personalized manufacturing, a core idea of Industry 4.0 proposed in 2011, envisions a future where products are tailored to individual needs. Could we see a parallel in personalized software?
AI is reshaping software in ways that challenge the traditional SaaS model, as venture capitalist Murray Newlands points out. He notes that AI agents are rapidly automating complex workflows and delivering hyper-personalized solutions. I take this further: with AI advancements, the barriers that once kept individuals from developing software are rapidly fading.
“Genius is one percent inspiration and ninety-nine percent perspiration,” Thomas Edison said. I wonder if AI could handle the "99% perspiration" in programming, freeing people to focus on the "1% inspiration." But this is more than just about perseverance — it’s about purpose.
Cognitive scientist John Vervaeke describes humans as autopoietic systems: we don’t just process information; we care about the information we process, while AI — at least for now — doesn’t. Whether AI could ever be “conscious” is outside the scope explored here. As AI becomes more adept at procedural problem-solving, we are still the ones deciding which problems matter. That crucial 1% is about problem-finding and deciding what we want for ourselves.
Extending Our Minds: The Cognitive Loop with AI
This transformation aligns with philosopher Andy Clark's concept of extended cognition — the idea that our tools become part of our thinking processes. AI doesn’t just automate tasks; it forms a cognitive loop with the human brain, amplifying our capacity to build, iterate, and innovate.
Isaac Newton said, “If I have seen further, it is by standing on the shoulders of giants.” But with the growth of human knowledge, climbing from foothills to the giant’s shoulders is increasingly difficult. AI may be able to help, by lifting us onto those shoulders.
Future Scenarios: Where Could This Lead?
A "Ridiculous" Future Without Software Companies
Renowned futurist Jim Dator famously said, “Any useful statement about the future should at first seem ridiculous.” So let’s entertain a seemingly absurd idea: a future where traditional software companies become obsolete. Code, like healthcare and education, becomes a public good. Individuals pull from a collective human knowledge database, aided by AI, to build just the tools they need. In this scenario, market incentive structures are completely undermined, and control over how people choose to use technology is returned to their own hands.
End of Big Tech Monopoly (But Not End of Big Tech)
A more plausible scenario might see Big Tech adapting to coexist. Just like YouTubers have not replaced Hollywood, Big Tech could focus on infrastructure, large-scale data management, and specialized services that require massive resources, while individuals and small groups create custom solutions for specific needs.
Challenges and Ethical Considerations
Of course, this democratization isn’t without risks. The “inherent danger of knowing just enough” means individuals might create tools without understanding potential security vulnerabilities or ethical implications. Education and guidelines will be crucial. We’ll need foundational knowledge in responsible software creation.
Conclusion
Empowered by AI, we’re approaching a future where the act of creating software could be as accessible as using it. It’s about redefining our relationship with technology — shifting from passive users to active creators. As we stand at this crossroads, the question isn’t just about what technology can do but what we want it to enable us to become.
In the late 2000s, Apple ushered in the era of ubiquitous personal computing with their campaign, “There’s an app for that.” Perhaps the natural next phase is, “I can make an app for that!”
Reference:
Clark, A. (2023). The Experience Machine: How Our Minds Predict and Shape Reality. Pantheon.
Newlands, M. (2023). AI Agents Are Eating SaaS. https://murraynewlands.substack.com/p/ai-agents-are-eating-saas
Vervaeke, J. (2019). Awakening from the Meaning Crisis. YouTube Lecture Series.
© Wensupu Yang, 2024
Wensupu (Wen) Yang, originally from China and now based in London, UK, earned a Bachelor of Science in Management Science from UC San Diego in 2020. After graduation, Wen discovered the field of foresight and instantly recognized it as his true passion. His self-directed learning journey in foresight includes completing the Foresight Essentials Training with the Institute for the Future (IFTF) and joining the Association of Professional Futurists (APF) in 2022. Now firmly committed to establishing a career as a professional futurist, Wen is driven by a passion for lifelong learning and an interdisciplinary approach to help organizations anticipate and adapt to complex futures.
Comentarios