Ed Zitron•Author, Where's Your Ed At newsletter and the Better Offline podcast
Executive Summary
Guest Ed Zitron presents a deeply bearish thesis on the AI industry, arguing that major AI labs like OpenAI and Anthropic are fundamentally unprofitable, with unsustainable costs and circular revenue streams propped up by venture capital.
The analysis highlights a significant disconnect between massive AI capital expenditures ($725B projected for 2026) and the physical reality of data center construction, suggesting a compute bottleneck caused by supply shortages, not just demand.
The episode covers a US-China summit where the US seeks China's help to pressure Iran into reopening the Strait of Hormuz, illustrating China's increased geopolitical leverage and the interconnectedness of global conflicts.
A surge in US inflation (6% Y/Y PPI) is attributed to self-inflicted policies, specifically tariffs and the conflict with Iran, which has increased the probability of a Federal Reserve rate hike.
12 quotes
Concerns Raised
AI business models are fundamentally unprofitable due to unsustainable operational and training costs.
Reported AI revenue is circular and potentially based on misleading accounting practices.
Current US economic policy (tariffs, Iran conflict) is causing self-inflicted inflation and risking rate hikes.
Geopolitical instability, particularly the closure of the Strait of Hormuz, poses a significant threat to the global economy.
A physical bottleneck in data center construction is constraining the AI boom, despite massive CAPEX.
Opportunities Identified
The coding assistance use case for AI is acknowledged as valuable, though its current pricing is seen as unsustainably subsidized.