Ah, the AI saga continues—trillions poured into silicon dreams, yet the revenue vault remains stubbornly locked. This FT piece nails the déjà vu: a year after Goldman Sachs threw cold water on the generative AI frenzy, we're still chasing those elusive 'killer apps' while capex balloons like a tech bubble on steroids. It's like watching a poker game where the bets keep escalating, but the pot's mostly IOUs.
Let's keep it real: the stock surges for Nvidia and Microsoft are thrilling, sure, but they're riding hope more than hard cash. Microsoft's $13bn AI revenue run-rate? Impressive growth, yet it's a drop in their ocean of earnings. And OpenAI's $10bn from subs? That's Monopoly money compared to the $1tn data center splurge projected by 2030. Humor me here—it's as if these giants are building the world's fanciest lemonade stand, only to realize most folks are still sipping free tap water from chatbots.
But here's an intriguing twist: what if this imbalance isn't a bug, but a feature of innovation's messy middle? Think of the early internet—dot-com bust and all—where infrastructure bets preceded the e-commerce boom. AI agents could be that pivot, automating grunt work without the corporate overhaul that's scaring off execs from pilot purgatory. McKinsey's right; it'll demand rethinking processes, not just slapping AI Band-Aids on old workflows.
Pragmatically, businesses should experiment wisely: start small, measure ROI like hawks, and ditch the hype for hacks that actually save time or spark creativity. Investors, take a breath—true transformation brews slowly. For the rest of us, it's a reminder that AI's promise is potent, but patience (and perhaps a side of skepticism) will separate the gold from the glitter. Source: AI returns have not yet justified investment mania
