The technology of our time. We believe AI represents the most transformative technological advancement since electricity. The similarities start in the way each improves the efficiency of past innovations. At the start of the 20th century, electricity turned mechanical machines into digital devices, Model Ts into the cars we have today and gas lamps into a light switch. Today, AI is turning those digital devices into personal assistants, cars into always-at-the-ready chauffeurs and is removing the need for light switches altogether – with everything from lights to music to movies on demand just by saying “Alexa”. But the real power of AI is what’s yet to come. Just as today’s personal computer was unimaginable in the 19th century, so too will be the innovations stemming from AI over the coming decades. Today, electricity is a simple utility. Similarly, AI is growing into the utility of tomorrow – with the monthly “compute” bill likely to sit alongside (or perhaps even replacing) the electric bill.
Great power, great responsibility. Policymakers’ ability to harness the tremendous power of AI will dictate its ultimate economic usefulness. Too much regulation will suffocate the burgeoning technology. Too little regulation could lead to major societal harm. These fears have always led to a backlash against new technologies – going back to the Luddites during the First Industrial Revolution (and likely before). This Fourth Industrial Revolution will face similar blowback and will require a well-reasoned approach. The answer is not to outlaw it altogether – as, if we don’t do it, someone else will. So, the show must go on, but it seems likely that getting to the correct regulatory balance will be an iterative process that plays out over years (and further refined over decades), with possible hiccups along the way. Overall, adoption disruption represents one of the key risks to AI investing.
For it to come, you need to build it. The other key risk is the massive capital expenditures that AI requires – semiconductor mass production, data center buildout, electrical-grid upgrades, etc. – alongside the needed raw materials to build it and the energy sources to power it. This will take a lot of money and a lot of time – and all needs to be done without impeding existing economic functioning. As such, the AI buildout will likely face periodic bottlenecks and inflationary challenges – especially in the ramp-up period of the next decade.
Risks but also potential reward. These risks are real and are reviewed in this report – including the size and scope of the buildout and the necessary regulatory considerations as AI begins to impact day-to-day life. But so are the investment opportunities – which are also reviewed in this report – across raw material/energy inputs, data centers and infrastructure, semiconductors, the hyperscalers (those creating the Large Language Models that make AI possible) and applications (the hardware and software that benefit from AI most).
Search
Personal
Wealth
