Automation Led to Economic Misery. AI Doesn’t Have To.

The relentless march of technological progress has long been touted as a panacea for economic woes, but the reality is far more complex. Four decades of automation have yielded unprecedented productivity growth – yet it has also left a trail of devastation in its wake. As we stand at the cusp of another revolution with the advent of generative AI, it’s imperative to confront the dark side of innovation and ask: what’s next?

Economic growth may have continued, but it’s been concentrated among a select few, while millions have seen their living standards stagnate or decline. This is not merely a consequence of automation; it’s a deliberate choice that has prioritized profit over people. The digital economy has created new winners and losers, with the latter often forced to bear the brunt of technological disruption.

Now, as AI gains traction, there are already signs that history may repeat itself. The focus on automation, rather than augmentation, threatens to exacerbate existing inequalities. Generative AI offers a different path forward – one that envisions technology not as a force for displacement, but as a tool for empowerment.

The idea of “pro-worker AI” has its roots in the 1960s, when computer scientist J.C.R. Licklider first predicted that computers would augment human cognition, providing workers with highly processed information to enhance their productivity. Decades later, we finally have the capability to deliver on Licklider’s prophecy.

Electricians could benefit from AI tools that draw on targeted knowledge bases and past use cases to help them troubleshoot complex equipment. In education, models trained on relevant curricular material could enable teachers to tailor lesson plans and spot patterns in test results. Even in healthcare, specially trained AI could enable lower-skilled laborers to perform a broader range of tasks.

But realizing these benefits will require a fundamental shift in how society approaches AI development. We must redirect our focus away from automation and toward augmentation – creating tools that expand workers’ capabilities, rather than replacing them. This won’t happen overnight; it’ll demand a collective effort from governments, corporations, investors, and workers alike.

One critical step is to rethink the way we fund AI research. Currently, most investment chases automation, driven by the promise of cost savings and increased profit margins. But this approach perpetuates the status quo, prioritizing shareholder interests over human well-being. It’s time to create incentives for pro-worker AI – through grants, public competitions, or other mechanisms.

The tax code also plays a significant role in shaping our economic landscape. Labor income is taxed at a far higher rate than capital income, discouraging businesses from investing in workers and perpetuating the cycle of inequality. Reforming the tax code to level the playing field could be a crucial step toward redirecting AI development toward pro-worker goals.

As we move forward, it’s essential to recognize that AI is not a panacea for economic woes – nor is it a zero-sum game where some must lose so others may gain. By embracing pro-worker AI, we can create a future where technology uplifts, rather than displaces, workers. The choice is ours; let’s choose wisely.