“Nobody knows anything” and “this time is different”: the phrases shaping — and shadowing — the AI economy

(SeaPRwire) –   Two phrases consistently emerge during every significant financial bubble in modern history.

The first is “this time is different,” which Sir John Templeton identified as “among the four most costly words in the annals of investing.” It signifies investors rationalizing inflated valuations by believing traditional metrics are no longer applicable. Carmen Reinhart and Kenneth Rogoff lent academic credibility to this notion in their influential 2009 book, This Time Is Different: Eight Centuries of Financial Folly, detailing how governments and investors repeatedly convinced themselves that past crises would not repeat, only to be proven wrong.

The second phrase, though less frequently discussed, is equally telling: “nobody knows anything.” This sentiment arises when uncertainty is so profound it leads to near paralysis.

As of May 2026, both phrases are prevalent, often voiced by the same individuals.

Ethan Mollick, a Wharton professor and a leading voice on AI, stated, “Nobody knows anything.” Speaking to hundreds of corporate leaders at the New York Public Library on Thursday morning, he explained, “I spend my time talking to AI labs, famous people, I talk to CEOs all the time, and nobody knows anything. We’re all making this up as we go along. So anyone who’s like, ‘We have the playbook’ — they’re lying to you.”

The Number Behind the Noise

Consider the figure that contextualizes all this: 0.1%.

This is Bank of America’s estimate for the current annual boost to economy-wide productivity attributed to AI, published in a report that also posited AI’s impact as greater than that of electricity and the internet combined.

Similarly, in March, Goldman Sachs reported “no meaningful relationship between AI and productivity at the economy-wide level,” while simultaneously noting a median 30% productivity increase in customer support and software, the two sectors where AI has seen the most concentrated application.

The calculation behind the 0.1% is straightforward. AI currently has the potential to transform approximately 20% of all workplace tasks. Of these, only 23% are cost-effective to automate at current prices. Automated tasks yield savings of roughly 27% in labor costs. Labor constitutes about half of all expenses. Multiplying these figures suggests a theoretical maximum gain of 0.66% in labor productivity today, before accounting for friction, delays, and institutional inertia that further reduce the realized number.

This is Bank of America’s own calculation, used to support its optimistic outlook. Every serious discussion about AI’s economic future, whether bullish or bearish, centers on whether, how quickly, and at whose expense this gap will close. The following presents the two most compelling arguments for each perspective.

The man who killed the playbook

Mollick’s assertion that “nobody knows anything” suggests, in a way, that the tech industry is in its Hollywood phase. This same sentiment was famously articulated by William Goldman, widely regarded as one of the greatest screenwriters, in his 1983 memoir, Adventures in the Screen Trade:

“Nobody knows anything,” concluded the writer behind Butch Cassidy and the Sundance Kid, All the President’s Men, The Princess Bride, and many other acclaimed films, who also won two Oscars. “Not one person in the entire motion picture field knows for a certainty what’s going to work. Every time out it’s a guess — and, if you’re lucky, an educated one.”

Mollick’s address to the New York Public Library audience echoed Goldman’s sentiment. “There’s no playbook,” he stated. “We’re figuring it out. On one hand, that’s terrifying. On the other, it’s great — because that means if you create your own playbook, there’s actually a source of advantage for you in that.” This contrasts with Hollywood studios that struggled to avoid producing box office duds.

The stock market crash mirrors the box office bomb, and Mollick indicated that outcomes depend on two simple yet challenging questions: “The biggest picture, there’s only two questions that actually matter a lot, which is how good and how fast? How long does this exponential curve continue and at what point does it ease off and how sharp will it be? That determines everything else.”

Current discussions are based on an understanding of the present, assuming the future will mirror the past, Mollick added. However, the “jagged frontier” of AI advancement, a term he popularized, makes this assumption inherently unstable. Yet, both Templeton and Mollick cannot be entirely correct; the playbook must still hold some relevance, or indeed, this time is truly different.

Mollick identifies organizational factors, rather than technological ones, as the primary reason for the delayed economic payoff of the AI curve. His explanation is more precise than the analyses provided by banks. He referenced an article he wrote for The Economist, suggesting that IT departments are where “AI goes to die”—not due to malice, but because their risk-reduction mandates are fundamentally at odds with experimentation.

“KPIs are the biggest enemy at this point. They force you into very bad paths in the experimentation phase,” he explained. “The very nature of saying we need a 10% improvement constrains the kind of use cases that you see.”

Transformative AI applications—those that replace existing processes rather than merely improving them—cannot be generated through KPI-driven approaches. This represents the organizational dimension of the 0.1% problem: not irrational exuberance, but rather rational conservatism embedded within quarterly earnings reports and performance review cycles.

The most compelling evidence that “nobody knows anything,” even more so than the 0.1% figure, comes from Mollick’s observation about AI companies themselves. “It’s weird that the AI companies are all now building their own consulting arms to do AI deployment. If the models are so good that you think they’re going to destroy all white-collar jobs, shouldn’t they also be able to help you deploy systems?”

The companies that developed the technology and are most optimistic about its potential are unable to use that same technology to answer the fundamental practical question: how is it actually deployed?

Depending on the perspective, this indicates either a lack of a playbook for current developments or the oldest playbook in existence: a new player has arrived, offering something everyone else needs.