Can Artificial Intelligence Be Governed—Or Will It Govern Us?

Key Takeaways

• Bush’s 1945 report built the NSF, a model for tech governance
• AI’s dual‑use nature mirrors early internet benefits and threats
• Institutional design, not just tech, determines AI’s societal impact
• Without governance, AI could dictate decisions, eroding human agency
• Collective policy action needed to align AI with public good

Pulse Analysis

Vannevar Bush’s 1945 essay "As We May Think" introduced the memex, a precursor to today’s internet, and warned of both its transformative potential and its dangers. By championing a networked memory for individuals, Bush anticipated modern AI‑driven search, telemedicine, and legal research. Yet his foresight also highlighted information overload and the weaponization of technology—issues that now surface in debates over large language models and autonomous systems. Understanding this historical lens underscores that technology’s trajectory is shaped not merely by invention but by the institutions that steward it.

In the current AI boom, the absence of clear governance frameworks mirrors the early internet’s wild west era. Companies race to deploy generative models while regulators scramble to define accountability, bias mitigation, and safety standards. The article argues that, as with Bush’s post‑war push for the National Science Foundation, deliberate policy structures are essential to channel AI’s benefits toward the common good. Nations are experimenting with AI oversight bodies, funding research on trustworthy AI, and drafting international norms, but fragmented approaches risk creating a patchwork that hampers innovation and leaves critical gaps.

The stakes are high: unchecked AI could dictate economic outcomes, influence elections, and automate critical decisions without human oversight. To avoid a future where AI governs us, policymakers must craft coordinated, transparent institutions that balance innovation with risk management. This includes public‑private partnerships, cross‑border standards, and investment in AI literacy. By learning from Bush’s legacy—where strategic institutional design propelled America to technological leadership—today’s leaders can build a governance ecosystem that ensures AI serves humanity rather than subjugates it.