HYPE around generative AI reached fever pitch in 2025.  Early 2026 tells a more sobering story as we witness AI's descent into the trough of disillusionment.

All innovative technologies go through this cycle, AI proving no exception.  Strap in as the tide of negativity is pulled ashore.

• At an industry level the descent is in full swing.  No major western AI model company is profitable, quite the contrary, they're burning piles of cash.  The math doesn't add up. Financial backers, falling over themselves in 2025 to fund all things AI, are now reserved and getting nervous, pulling funding from AI infrastructure build-outs.  The AI bubble dominated headlines as 2025 closed, analysts now speculate on what the deflation will look like.  The only matter of debate is quite how fast, deep and wide the correction will be.  The skeptics told us so.
• At a business level the misguided expectations of massive labour savings and paradigm-busting transformations proved largely elusive, underwhelming, or for some, unfortunate fantasy.  Countless failed AI technology deployments were tolerated as "the price of innovation" in 2025, some now seem laughable.  Business cases for many big-bet AI investments come off retrospectively as highly questionable, "just add AI" gained board approval and financial backing, and surprise surprise, mostly failed to deliver.  What's ROI again?
• At a collective level negative perceptions have firmly taken hold, bad press far outweighing positive press in what little we've seen of 2026.  No longer the darling, now the punching bag of the media cycle, the overall sentiment toward AI is now one of financial and environmental recklessness, and failure to live up to hype.  Yesterday's stories of wonder, promise and revolution are but a memory, AI now more synonymous with slop than superintelligence.

It’s all a little grim! But good news is that nothing has fundamentally changed, this is a perception shift. And this brings a useful balance and grounding to the discussion, dispelling unhelpful industry-perpetuated myths, shining much-needed light on the topic. All signs that the direction of travel is towards maturation.

Those that took risks in 2025, invested and experimented with AI projects may have experienced the bumps and scrapes of short term failure, learning the hard way about what doesn't work, but gained AI maturity and stepped closer to constructive AI adoption.  Largely the failures were not due to AI technology itself, but instead were rooted in naive deployments, where culture, data, leadership and organisational change were lacking.  Those that took the less flashy route of building human augmentation have largely circumvented the hype/crash cycle and steadily scaled their capability as a business, wielding today's tools innovatively.

In 2025 the biggest risk for business leaders was being swept up in ungrounded promise, making regrettable misguided investments.  In Q1 2026 the risk is becoming completely disillusioned and bogged down by dominant negative sentiment.  For those that persist, the best is yet to come.  Like previous technological revolutions, the most impactful changes that AI will bring about aren't even apparent to us yet.  They will unfold long after this short and volatile perception cycle.  Glimpses of the "slope of enlightenment" are visible to those able to peer through the dim pessimistic fog.

Pro tip: Don't let negativity and skeptical dogma shut down balanced discussion and decision-making around AI. Success stories, both transformational and humble, are happening every day, not that you'll hear about them for the next few media cycles!

Projection: 2026 Q1 brings continued descent into the trough of disillusionment while AI remains a media punching bag.  Q2 sees the perception of AI hits rock bottom.