🤖🎢 Where are we in the AI 'hype cycle?'

Despite some rising skepticism, the current frenzy may be laying the groundwork for a transformative technological revolution

Back in February 2020, it was profoundly weird — certainly in retrospect, but even at the time, really — that the Democratic presidential debate had nary a mention of the obviously exploding COVID-19 pandemic. I think there’s a good chance that last week’s presidential debate will also prove notable for what it didn’t mention: the emerging AI Revolution. So many questions that could have been asked:

• “What can government do to make sure US companies take full advantage of AI and that the technology continues to rapidly advance?”

• “Washington took a light-touch regulatory approach with the internet in the 1990s. Why not do the same with AI?”

• “Given the potential for rapid AI advancement to human-level capabilities and beyond this decade, should the US government take immediate steps to secure and nationalize AI research and infrastructure, and if so, what specific actions should be prioritized?”

Alas, so many missed chances. But you can’t blame just the CNN moderators for neglecting the topic. Joe Biden issued a sweeping executive order on artificial intelligence last October. Donald Trump gave his views on the issue during a recent podcast with venture capitalist David Sacks. Both candidates had ample opportunity and rhetorical ammo to address the potential risks and rewards of recent AI advances as they outlined their vision for America’s future.

Outside the Washington media bubble, AI is a pretty big deal — especially since the November 2022 rollout of ChatGPT. You can see it both in corporate investment and the rising share prices of AI-related companies. Chipmaker Nvidia is probably the best example of both trends. We’re still waiting, however, for solid signs that AI ( especially the large language models of generative AI) is boosting US productivity growth. Still, there are strong expectation by many economists, technologists, and CEOs that it will do so eventually in a significant way.

For example: Capital Economics thinks AI is beginning to boost business investment in the US, though evidence is still somewhat limited. Eventually the firm expects AI investment to significantly increase productivity in the US, potentially by 1.5 percentage points annually (an impact likely to materialize only in the latter half of this decade). From a new CE report: “To this end, we are not particularly concerned that there is not yet much evidence of AI having transformed productivity growth.”

That said, the disconnect between investor/CEO enthusiasm and significant economic impact can be seen as a classic case of a “hype cycle” in action, specifically the famous Gartner Hype Cycle. It’s a descriptive model that illustrates how perceptions of emerging technologies evolve. In the case of AI, we seem to be nearing the "Peak of Inflated Expectations,"suggesting an impending period of skepticism before the technology's true benefits materialize.

Capital Economics