✳flâneur — a map of the web's best reading

'The AI Dilemma: Growth vs Existential Risk': An Extension for EAs and a Summary for Non-economists — EA Forum

forum.effectivealtruism.org · 246 words · saved by 1 readers

Update (14 Aug 2025): The code generating results in Table 4 (optimal pause decisions) contained an error leading to local rather than global optimization. I have left Table 4 in it's original form, but new code can be found here which reveals optimal non-zero pauses for ρ = 2 % , A = 0.2 and γ > 1.01 . In this post I first summarize a recent paper by Chad Jones focused on the decision to deploy AI that has the potential to increase both economic growth and existential risk (section 1). Jones offers some simple insights which I think could be interesting for effective altruists and may be influential for how policy-makers think about trade-offs related to existential risk. I then consider some extensions which make the problem more realistic, but more complicated (section 2). These extensions include the possibility of pausing deployment of advanced AI to work on AI safety, as well allowing for the possibility of economic growth outside of deployment of AI (I show this weakens th

SummaryBot 2y 1 0 0 Executive summary: The decision to deploy AI that could increase both economic growth and existential risk depends critically on the potential growth benefits, the existential threat, and the curvature of the utility function, with extensions showing that economic growth outside of AI deployment and the ability to pause AI to reduce risk can significantly impact optimal deployment decisions. Key points: The decision to deploy AI involves a trade-off between potential unprecedented economic growth and increased existential risk. The optimal AI deployment time is mainly deter

Explore this link on the map →

related reading