Scarcity Captures the Mind

This is a submission for the 2026 WeCoded Challenge: Echoes of Experience

“Scarcity captures the mind. […] The mind orients automatically, powerfully, toward unfulfilled needs.”

— Sendhil Mullainathan and Eldar Shafir, Scarcity: Why Having Too Little Means So Much

In the world of AI and software, this often means not having enough computer power—like RAM, graphics cards, cloud credits, or even the ability to run models on your own computer without it slowing down or crashing. When these resources are scarce, your equipment suffers as well as your thinking. You feel more cautious, less willing to take risks, and your innovative ideas slow down because it’s so hard to try things out.

Think about many developers, especially those from underrepresented groups in tech. You might get excited to work with open-source AI models at an event, but your old laptop with only 8 GB of RAM struggles to handle the task. The smallest AI models crash immediately, and trying to fine-tune models on free platforms becomes too expensive. Because of this, you stick to simpler projects, avoid more demanding tasks, and hold back on exploring your ideas. The lack of resources drains your mental energy and motivation. You end up spending more time worrying about what you don’t have than focusing on what you could create.

Mullainathan and Shafir’s research shows that scarcity doesn’t just limit your tools — it limits your bandwidth.

But when you finally get better hardware or enough computer power, everything changes. Suddenly, you’re more willing to try new things, experiment with more complex models, explore ideas that seemed impossible before, and be more creative. The tools are there, just waiting for you, boosting your confidence and encouraging innovation. It’s amazing how access to better resources can unlock your potential and make ideas grow faster.

Prominent voices in AI have echoed this dynamic for years. Andrew has repeatedly described AI as “the new electricity” — a transformative force that democratizes possibility, yet only for those who can plug in. Andrej Karpathy, in his teaching and writing, stresses hands-on experimentation as the fastest path to deep understanding in deep learning; that experimentation requires compute most early-career or resource-constrained developers simply don’t have. Fei-Fei Li’s pioneering work in computer vision and her advocacy for human-centered AI underscore the same point: breakthrough ideas flourish when the infrastructure exists to test them at scale.

On X, the conversation is equally pointed. As one recent post highlighted amid broader discussions on AI’s future:

“Only 27% of women are on AI teams. Only 25% minority representation. 29% of AI teams have zero minority employees… We’re building ‘intelligent’ systems while excluding traditional knowledge systems that have successfully guided human decisions for thousands of years.”

These numbers aren’t accidents of interest. They reflect systemic resource gaps — economic, educational, and institutional — that hit women and marginalized developers hardest. When compute is gated behind expensive hardware or paid APIs, entire communities are quietly locked out of the very experimentation that fuels innovation.

Sheryl Sandberg’s Lean In and Reshma Saujani’s work with Girls Who Code make this connection explicit in the broader fight for gender equity. Support networks and education matter, but so does raw access to the tools. Without them, talented developers are forced to play defense instead of offense.

When barriers fall and resources become available, creativity floods. Diverse teams with equitable access redefine what’s possible. Ideas that once died in low-RAM laptops now scale into production. Concepts that felt abstract suddenly become prototypes. Growth becomes exponential, not incremental.

In short, making sure everyone has equal access to powerful technology is at the heart of supporting gender equity and diversity in tech. It’s about providing access to tools like community labs, affordable cloud credits, open hardware projects, and programs that treat access not as a privilege but as essential infrastructure. When all developers—no matter their gender, background, or where they live can experiment freely and confidently, we unlock a wave of human creativity that AI was meant to amplify.

Yes, resource limitations are real. But the moment those limits are minimized, the potential for innovation roars to life. Let’s work toward a future where no one has to wait to dream bigger because they lack the necessary tools today.