What I read this week

8 stories that caught my eye this week and what I took from them.

Could Advanced AI Drive Explosive Economic Growth? – Tom Davidson

• Could Advanced AI Drive Explosive Economic Growth? | Open Philanthropy
• Could GDP growth accelerate by 10x, to 30% per year?
• Very big picture, very long term, but a fascinating thought experiment.
• Zooming out from our own frame of reference, and the period we have the best data for, growth rates have increased significantly over history. If you knew nothing about economics, would it not be strange to assume that this process has stopped?
• My favourite quote: “It took thousands of years for growth to increase from 0.03% to 0.3%, but only a few hundred years for it to increase from 0.3% to 3%. If you naively extrapolate this trend, you predict that growth will increase again from 3 percent to 30 percent within a few decades.”
• The ideas feedback loop argues: more ideas -> more output -> more people -> more ideas. But over the past century this has been broken and birth rates have declined with greater wealth. This has been the period where growth rates have been relatively steady. What if AI revives this loop: more ideas -> more AI -> more ideas?
• If AI systems replace human labour, a part of the growth equation goes exponential.
• “Could hypothetical economists a thousand years ago, who saw the economy grow at a snail’s pace, have imagined how the processes of industrialization and technological innovation would allow the modern economy to double every 30 years?”
• Caveat: For AI to drive explosive growth, AI systems would have to be capable enough to replace human workers in most jobs, including cutting-edge scientific research, starting new businesses, and running and upgrading factories.
• The Metaculus community predictions is that this level of ‘transformative AI’ will be reached in a little over a decade (2036), so maybe not as far off than you would have thought.
• Bottom line: too long-term for any tactical views, but a fascinating thought experiment that challenges our recency bias. And it’s a big deal for asset prices if only a fraction of this plays out or the Metaculus timeline is correct.

EV Battery Costs - Dropping Faster Than Predicted Even By Tony Seba – TorqueNews

• EV Battery Costs - Dropping Faster Than Predicted Even By Tony Seba | Torque News
• Li-ion batteries dropping to $56/kWh…from >$500/kWh a decade ago
• Battery prices continue to follow the multi-decade trend of price declines. If anything deflation has accelerated recently.
• Bottom line: Good news for decarbonisation. The cultural debate may be becoming more muddled, but the technologies continue to progress. Ultimately, economics will decide the debate. A broader question: what changes if the marginal cost of energy goes to zero?

‘The Right Kind of Crazy’: Matt Gaetz Makes MAGA World Swoon – Politico

• ‘The Right Kind of Crazy’: Matt Gaetz Makes MAGA World Swoon - POLITICO
• Maybe it’s too early to think about what happens ‘after Trump’. Even if he loses in November, he could well run again in 2028.
• But the last few months have shown that Trump, and also his style of politics, have an enduring appeal to American voters, at least to the base of the Republican party.
• This story is about Matt Gaetz, but the success of Vivek Ramaswamy in the primaries highlights a similar pattern: ‘MAGA style’ is here to stay.
• Bottom line: Donald Trump is a symptom, not the cause. After Trump, expect more Trump, even if it has a different face.

10 Breakthrough Technologies 2024 – MIT Technology Review

• 10 Breakthrough Technologies 2024 | MIT Technology Review
• AI, obviously, but the more interesting bit is what else is on the list that has received less attention than it deserves because we are all staring at all things AI.
• 3 in Clean Tech, 2 in Healthcare, 2 in computing
• One that caught my eye: CRISPR. CRISPR has been very promising for years, but the first treatment getting approval in the US and the UK is a milestone for the technology. There are more in the pipeline.
• Bottom line: technological progress continues outside of AI as well

AGI timeline – Metaculus

• Date Weakly General AI is Publicly Known | Metaculus
• Date of Artificial General Intelligence | Metaculus
• None of us can predict the future, but prediction markets and prediction experts are one of the best approaches to thinking about outcomes.
• Metaculus has a few fascinating questions on when we might achieve artificial general intelligence.
• Predictions are for weak AGI (75^th percentile of human abilities across subjects) in 2026 and ‘proper’ AGI in 2031.
• My take-aways from looking at the predictions were 1) that’s sooner than I thought; 2) timelines have accelerated, but not a lot, since the ChatGPT moment; and 3) timelines have not been pushed out with the passage of time, as we often see with recession forecasts.

Your Healthspan Is as Important as Your Lifespan—and It’s Declining – Wall Street Journal

• Americans Are Sick for More of Their Lives - WSJ
• The Lifespan-Healthspan gap is growing, and that’s not a good thing
• Lifespan = total number of years lived. Healthspan = number of disease-free years lived.
• This adds to the decline in life expectancy, especially in the US.
• Living a greater share of your life with diseases has economic implications via the costs of the healthcare system and societal implications in the decline in ‘population sentiment’ which feeds into the rise in populism.
• Does this trend add to the structural demand for GLP-1 drugs?
• Bottom line: get some exercise, get your 5-a-day and do your bit to reverse the trend, at least on a personal level

I spent the morning with the Apple Vision Pro – TechCrunch

• I spent the morning with the Apple Vision Pro | TechCrunch
• Pre-orders sold out quickly, but at $3,500 the Vision Pro will be a niche product for some time.
• The reviews generally sound impressive, though, so give it a few iterations and maybe there’s a real market there.
• Bottom line: we’ve all but forgotten about the Metaverse, but fundamentally this slow-burning theme continues to develop. One to watch for the next stage of the hype cycle: the ‘slope of enlightenment’

Innovation GPT - One Useful Thing

• https://chat.openai.com/g/g-JaiQEuHRU-innovator
• Ethan Mollick has put together a great (free) innovation tool, based on GPT-4.
• It goes through convergent and divergent thinking processes and gives you a list of ideas in just a few minutes. As so many things AI, not all are great, but some are things you had not thought about and can spark a new discussion. Worth a try.

Thanks for reading Macro Equity! Subscribe for free to receive new posts and support my work.