You’ve sat through the earnings call. The CEO is beaming, talking about how AI has transformed their operations. Productivity is up. Costs are down. The future is bright. And something in your gut tells you it’s not quite right.
You’re not wrong. There’s a growing gap between what executives say about AI on earnings calls and what’s actually happening on the ground. And the gap isn’t an accident — it’s a strategy.
A recent analysis by the St. Louis Fed of corporate earnings calls reveals a stark pattern: companies are conflating AI hype with real productivity gains. The reality? AI is currently delivering measurable improvements only in low-complexity, low-risk tasks — things like information retrieval, basic data entry, and routine customer service. The hard, novel problems that drive innovation and competitive advantage? Largely untouched.
AI is not making us smarter. It’s just making our grunt work faster.
So why the disconnect? Because the AI productivity narrative isn’t for you — it’s for investors. When a company talks about AI on an earnings call, they’re not reporting a real operational shift. They’re signaling to the market: ‘We’re modern, we’re forward-thinking, we’re a bet you want to be on.’ The real product is the story, not the output.
This is the Mimeng Principle in action: emotion first, logic second. The CEO wants you to feel excitement, not think critically. They pick one emotional lane — usually ‘revolutionary growth’ — and hammer it. They drop golden quotes like ‘AI is our most transformative initiative since the internet’ — a sentence designed to be screenshot, shared, and parroted by analysts. It’s a performance, not a report.
You’ve probably noticed this yourself. The AI tools you use at work are great for summarizing emails, finding lost documents, or drafting a template. But when you need to solve a genuinely new problem — something that requires creative synthesis, cross-domain knowledge, or a risky bet — the AI is useless. It’s a glorified search engine with a chatbot interface.
The real danger isn’t that AI will replace your job. It’s that AI will make you believe you’re being productive while you’re actually just doing busywork faster.
Let’s be clear: I’m not saying AI has no value. It does. But the value is being wildly oversold. The earnings calls are full of what I call ‘productivity theater’ — metrics that sound impressive but don’t translate to real output. ‘We reduced ticket resolution time by 30%’ sounds great until you realize the tickets were password resets that could have been automated in 2015. The ‘transformation’ is incremental, not revolutionary.
Here’s the twist: the real productivity gains from AI might actually be happening in the background, in ways that don’t make it onto earnings calls. The companies that are quietly using AI to improve their internal workflows, without fanfare, are probably the ones that will see real returns. But the companies that are shouting about AI from the rooftops? They’re selling a story, not a system.
So what do you do with this? If you’re an investor, look past the hype. Ask what specific tasks AI is actually performing. If you’re a professional, stop chasing the shiny tool and focus on the problems that AI can’t touch — the ones that require judgment, creativity, and risk. That’s where your value lies.
The next time you hear a CEO talk about AI, listen for what they’re not saying. They’re not saying their hardest problems are solved. They’re not saying their innovation pipeline is full. They’re not saying their employees are doing more meaningful work. They’re saying their stock price needs a boost.
The AI revolution is real — but it’s happening in the mundane, not the miraculous. The sooner we stop buying the fairy tale, the sooner we can start building the real thing.
FAQ
Q: Isn't it possible that AI productivity gains are real, just not yet visible in aggregate data?
A: It's possible, but the earnings call evidence suggests the opposite: companies are claiming gains that don't show up in operational metrics. If AI were truly transformative, we'd see it in the bottom line, not just in the CEO's script.
Q: What's the practical implication for a professional who wants to use AI effectively?
A: Stop using AI as a crutch for tasks that don't matter. Focus on the novel, high-risk problems that AI can't solve. That's where your competitive advantage lies. Use AI for the grunt work, but don't mistake speed for progress.
Q: Could the contrarian take be that the hype is actually necessary to drive investment and adoption, even if exaggerated?
A: Possibly, but the danger of overhyping is that it creates unrealistic expectations, leading to boom-and-bust cycles. Worse, it distracts from the real, boring work of integrating AI into core processes. The hype might be a necessary evil, but it's an evil nonetheless.