Was this the Fimbulwinter?

Norse mythology describes a severe, long winter that precedes the breakdown of social order and Ragnarök, the final battle among the gods. The losses are enormous, and the world will never be the same again.

At times this winter has felt a little like that.

The last remaining snowbanks have melted. Green grass has resurfaced, and the winter aconites and snowdrops are blooming again. That follows a season that was unusually persistent across Scandinavia and other parts of the Northern Hemisphere, including New York City, where temperatures had not been this low since the late twentieth century.

Spring invites renewal. Yet it also invites misjudgment.

Corporate executives should step forward with close attention to what has changed and remain conscious of how they react, while skylarks and blackbirds once again fill the air with their song.

The reason is simple: many executives are about to misread the meaning of artificial intelligence.

What’s forever altered?

A powerful force is at play and will be impossible to hold back. One proof point came when Jack Dorsey recently announced that his company, Block — the parent of the Square payment solution — would lay off 4,000 of its roughly 10,000 employees because of efficiency gains from artificial intelligence. This comes at a time when Block remains financially healthy.

The soft-spoken Vipassana meditation practitioner Dorsey stood in sharp contrast to the combative, serial job-cutter Elon Musk during their battle for power at Twitter and Musk’s subsequent takeover. Yet Dorsey became the first significant Silicon Valley CEO to openly link large layoffs solely to productivity gains enabled by AI.

The build-up to announcements like Dorsey’s has been long. But the momentum accelerated dramatically during the past winter. The shift is unfolding not in theory but among human beings of flesh and blood.

Over the past few months, the best software programmers have multiplied their productivity. They can accomplish tasks that previously required tens or even hundreds of people. Using tools such as ChatGPT, Claude, and Gemini, they have assembled virtual armies composed of hierarchies of agents commanding yet other agents. Many no longer write code line by line but instead “vibe code,” instructing layered systems of coding agents.

While the phenomenon is most pronounced in software programming, it requires little imagination to see how similar agent-based systems will conquer many other repetitive mental tasks not already addressed by enterprise software.

The consequences for companies and their executives — as well as for corporate professionals and society at large — are massive and irreversible.

Who should be particularly careful?

Spring itself invites impulsivity, as winter clothes are packed away, the sun climbs higher in the sky, and the cupids roam more freely. The risk of overreaction grows even larger when AI reaches an inflection point, producing an extraordinary productivity leap in one of the professions that until recently could never get enough of the sharpest human minds.

One group of executives is especially prone to overreacting: those who are not technology-native, unlike Dorsey and Musk.

Knee-jerking comes with surprise.

Non-techies are more likely to have been distracted. They have typically spent less of their lives in front of a screen behind drawn curtains. More than three years after ChatGPT became publicly available, the layperson’s initial excitement about AI has faded, whether or not “Chatty” has become a habitual part of daily life, much like internet browsers and search engines once did.

Dinner-party conversations about AI naturally revolve around whether the technology has created bubbles in certain tech stocks rather than the evolution of coding practices among programmers.

When those executives realize what is happening — which many should be doing about now — they often lack an intuitive feel for how technology works. Someone else usually handles the nitty-gritty details when their companies buy and implement software solutions or outsource operational processes.

The consequences of executives’ lack of technological intuition have already been visible in the recent management fashion around Agile ways of working. For many companies that followed the trend, it amounted largely to copying observable rituals such as daily scrums and Kanban board meetings. Yet the real impact has often been minimal when executives and middle managers failed to grasp the deeper philosophy behind the approach.

The Agile Manifesto of 2001 emerged from a grassroots movement among software developers who sought to challenge bureaucratic project management and instead emphasized self-organising teams.

Despite the popularity of agile methods in large corporations, most still run a constant stream of top-down projects. The size of their organizations may make some of this unavoidable. Yet fast-growing software-oriented companies have moved in a different direction. Instead of one team building software and another running it, they increasingly combine these responsibilities. The focus shifts from projects with fixed start and end dates to products that evolve over time. Cross-functional teams take responsibility across the full lifecycle, enabling faster learning and continuous improvement.

The advance of AI and the rise of layered agents will spark further innovations in how both creators and users work.

What is the blind spot?

A common overreaction to the rapid progress of AI will be an urge among executives to prove that they are not falling behind. Peer pressure will soon push them to issue bold statements about radical change and massive layoffs. Little will hold them back from spending substantial additional sums on top-down initiatives involving management consultants and software vendors.

History — and current developments — suggest that such approaches to technology carry high failure rates and disproportionate costs. There is no panacea for something as complex as running a large corporation in a dynamic marketplace within a rapidly changing world.

Irrespective of how ambitious a quest executives undertake in the AI space, they should pay attention to one vital element in creating a better future. The story of Fimbulwinter points precisely to that.

The Fimbulwinter was apocalyptic, but it was not the end. Odin and Thor were among the central gods who were killed. Humanity was almost erased, with only two exceptions: Líf and Lífþrasir, the mythological forebears of the Vikings.

There can be truth even in myths. Archaeologists believe that the climate catastrophe of 536 CE may have inspired the legend of Fimbulwinter. Massive volcanic eruptions darkened the sky and led to prolonged cooling, famine, and starvation across the Northern Hemisphere.

In the aftermath, Líf and Lífþrasir emerge into the light in the tenth-century poem Völuspá. The first name literally means “life”; the second is often interpreted as “life’s lover” or “life’s pursuer.”

Like the Vikings, the programmers pioneering vibe coding appeared one by one. The population mastering these new techniques spreads, much like winter aconites and snowdrops across a garden.

From a distance, AI may appear to be purely a money game. On closer inspection, its application has much to do with human curiosity, creativity, and continuous learning. Every master of a new art practices relentlessly.

That insight extends far beyond programming. Barely any imagination is required to see how repetitive human tasks will increasingly be automated. Yet efficiency and speed alone can never ensure the success or impact of a business.

Even the most overreacting executives must find an evolving balance between how they apply AI and how they treat people.

It is a paradox: executives who lack a feel for AI risk losing their touch with people — and, conversely, those who do not appreciate their own and others’ human nature will struggle to harness AI fruitfully.