AI Collaboration or AI Dependency?

I recently read an article about something called AI gravity (AI gravity – MIT Sloan). The main idea is that as AI becomes more capable, we naturally feel pulled toward outsourcing more and more of our thinking to it. The article argues that while AI makes us more efficient, it also raises an important question: what happens if we stop exercising the very skills that make us creative and capable in the first place?

Over the past year, I’ve interviewed public-sector employees about how they use AI in creative and innovation work. Some interviewees admitted they had become aware of relying on AI more than they intended. Others worried that summaries generated by AI could omit important nuances or that constantly asking AI first might slowly weaken their own ability to think through problems independently.

The easier AI made certain tasks, the more tempting it became to hand over the difficult parts of thinking. And ironically, those difficult moments (wrestling with uncertainty, refining ideas, questioning assumptions) are often where creativity actually develops.

Perhaps that’s the paradox of generative AI.

The more capable these tools become, the more valuable distinctly human capabilities become too: critical thinking, judgment, curiosity, communication, and knowing when not to accept the first answer.

AI may change how we work, but I don’t think it changes what makes great work. If anything, it makes those human qualities even more important.

That’s one of the biggest lessons I took away from both the article and my own research.