I sat down earlier this week with Cris Sheridan of Financial Sense to discuss my recent report on "The AI Trade". I broke down my concerns about enterprise adoption rates, unit economics, and buildout constraints. This year will see roughly $200 billion in enterprise AI spend. By best estimates, that number must reach $2.5 trillion to offset AI CAPEX, more than the tech sector's combined revenue currently. Enterprise AI usage is still plagued by model weaknesses, from hallucination to agentic token burn to a bias to please. Meanwhile, domestic and Chinese competition is compromising pricing power. And the buildout faces seemingly intractable supply chain, labor, and electricity limitations alongside an intensifying political backlash. You can read the executive summary of my report on "The AI Trade" here: sageroadresearch.com/product… You can listen to a few minutes of the interview here: financialsense.com/podcast/2…

WSJ: "The AI build-out is on track to become the biggest economic bet in US history, dwarfing investments made to fund other huge US infrastructure projects such as the railroads, the highway system and the plumbing for the internet. Investment in data centers and related AI infrastructure is projected to total $10.3t from 2025 to 2032, according to new estimates by economist Stijn van Nieuwerburgh. That is a staggering 3.6% of GDP a year, on average. Never before has the US economy been so dependent on the build-out of a single industry." Deeply concerning given how deeply AI spending is now woven into the US and global economy and how many threats remain that could derail expectations of AI ROI on any reasonable timeline. For just a snapshot overview of the AI vulnerability, here are some stats from my report on "The AI Trade" (sageroadresearch.com/product…): "To put some numerical context to AI bubble warnings, in our December report on “GenAI & Productivity”, we cited a calculation by former IMF Chief Economist Gita Gopinath: “A market correction of the same magnitude as the dotcom crash could wipe out over $20t in wealth for American households, equivalent to ~70% of American GDP in 2024.” She also calculated that foreign investors would lose ~$15t. She made those calculations in late 2025 and given how AI-trade concentration has increased since, undoubtedly those numbers are now greater. AI investment now accounts for 25% of US GDP growth and 8% of total GDP, exceeding the dotcom-bubble’s peak of 6.5% of GDP. The value of AI-linked firms has climbed ~$27t in the past three years, equivalent to 36% of the value of the entire US stock market. As a result, the five biggest companies in the S&P 500 now account for ~30% of the indexes’ total market cap, up seven percentage points versus five years ago. And AI-driven concentration is not just a US equity phenomenon—the top 10 firms in the MSCI Emerging Markets Index account for 41% of its total market cap, with just three semiconductor manufacturers making up 29%. Then there’s the debt picture: roughly a quarter of US gross investment-grade debt issuance this year has come from AI-related companies. And the venture capital picture: over the past 12 months, AI companies have received roughly two-thirds of all US venture capital dollars." History does not lend confidence about AI's path forward. As the FT wrote in late June, recapping findings of a BIS study: “Historical episodes of investment booms provide ‘instructive parallels’—among them the expansion of canals in the 1830s, railways in Britain in the 1840s and the dotcom boom of the late 1990s. These all had one key feature in common, a genuine technological breakthrough that attracted capital in excess of what commercial returns could ultimately justify.” WSJ link: wsj.com/economy/the-ai-build…