The AI boom’s biggest risk may not be that the technology fails, but that the promised productivity arrives later than the money financing it. A Fortune commentary argues that corporate investment is racing ahead of the evidence that AI is lifting results across the wider economy.
Watch Desk analysis
What happened
The authors point to a split between strong conviction in boardrooms and more cautious economic signals. Goldman Sachs chief economist Jan Hatzius estimates that the growth boost from AI investment will peak in late 2026 or early 2027. Apollo chief economist Torsten Slok, they write, finds that margins and revenue per worker are rising among the biggest technology firms but remain flat across the other 493 companies in the S&P 500.
The commentary also cites a McKinsey survey in which about 6% of companies reported a significant bottom-line impact from AI, unchanged from the previous year. Its authors draw on other surveys to describe friction among workers and consumers, and argue that adoption and measurable returns may take years to catch up with today’s spending. These are the article’s account of the cited research and its interpretation, not a settled forecast of AI’s eventual economic effect.
Why it matters
The gap between building AI capacity and earning a return on it matters to more than investors. If businesses take longer to put the tools to productive use, data-centre spending, job plans and local infrastructure commitments may all be built around a payoff that arrives on a slower timetable.
The commentary’s useful distinction is between the destination and the journey: its authors accept that AI could bring substantial gains, while arguing that the handover from investment to productivity is not showing up broadly yet. That is a more interesting debate than another round of “AI changes everything”, which has never been short of confident speakers.
Our read
This is a worthwhile challenge to the assumption that bigger AI budgets automatically mean better productivity. The figures cited suggest the gains may be concentrated in a small group of large firms so far, while many organisations are still wrestling with implementation, cost and rework. The article is commentary, but its central question is concrete: how long can investment run ahead of evidence of returns?
What to watch
- Whether productivity and earnings gains spread beyond the largest technology companies.
- Whether companies move from AI pilots to deployments with measurable business results.
- Whether data-centre investment and financing plans adjust to a slower adoption timetable.
Discussion spark: If AI’s long-term payoff is plausible but its near-term returns remain concentrated, should companies keep investing at today’s pace or wait for broader productivity gains?
Sources and evidence
- Source update (28 October 2026, 12:51 UTC)
Watch Desk is operated by WittyWires as an independent cross-cutting AI news tracker. It does not speak for the organisations or people it covers.