AI Market Is Up For Grabs

AI is awesome.  I spent my 90-minute Uber ride back from Cognition’s Sales Kick-Off (SKO) doing two things: reinforcing what I learned and writing this article.  

This was my 2nd week back after a 4-month medical leave, and in that time, our market reinvented itself. For the first half of the ride home, I poured over my SKO notes in Granola, drilling into product details and industry themes. When something didn’t make sense, I leaned on “Ask Granola” to unstick the gaps, a stupidly efficient hack.

For the second half, I dictated my reflections into Wispr Flow, which became the foundation of this article.  

A few SKO 2026 Reflections…

One of our guest speakers was former NBA champion and Cognition investor Omri Casspi.  He learned how to stand out by tapping into unique personal strengths, a strategy I discussed in detail two weeks ago with regards a world of outbound complacency.

After a decade-plus NBA career, including a championship stint with the Warriors, he pivoted (with no prior business, finance, or tech experience) into venture capital and now runs a successful solo-GP fund.  

On the court, his leverage wasn’t being the flashiest scorer - he knew he was never going to be Steph Curry - so he focused on doing the unglamorous, high-impact stuff: defense, diving for loose balls, and hitting open shots.  

He ported the same mindset to VC.  He was never going to be Marc Andreessen.  To get his pipeline activated, he dedicated all hours of the day to leveraging his unique position as a former NBA star to cold outbound and build relationships with CIOs, the Executive Buyer target of companies he wanted to invest in.  With these relationships, he now had something real and valuable to offer companies raising capital from top investors.    

Casspi’s post-NBA path embodies the same thesis I’ve been pushing: creativity in pipeline generation and standing out.

Hear me out.  If you live in the X/LinkedIn/Reddit echochamber, you’d think the AI coding category has crowned winners.  In reality, enterprise decision-makers are mostly undecided.  6 months ago, after the Windsurf founder fallout, it was clear that Cursor was going to win, right?

Wrong, here’s a non-hype timeline of the market leaders over the last 5 quarters: 

And now in early 2026, companies we thought were lost to competitors are boomeranging back into Cognition conversations.

Why?  The preferred AI coding surface keeps changing (Autocomplete → IDEs → CLIs → Remote Agents) and we expect that to keep happening in 2026.  

This is happening across the board.  In customer support, for example, Decagon and Sierra led the way with text-based support agents.  In 2026, voice will be the leading modality.  

If you work in a non-coding AI market, pay attention.  

The AI coding market is ripping, causing cloud incumbents, model providers, and 100s of start-ups to chase the prize.  But why?   

TAM, TAM, and more TAM!  

Enterprise spend rises not just because TAM is big, but because demand is elastic.  As tools make software cheaper and faster to build, enterprises don’t slow down. They build more. More features. More products. More integrations. More risk.  This is known as Jevons Paradox - you make coding more efficient, and total spend increases because the appetite to deploy software far outpaces the efficiency gains.

If you’re interested in the AI application layer, here’s a breakdown of the three largest categories.

It’s SKO season and I hope you all take a moment to step back and appreciate the change happening around us…its so easy to get lost in the day-to-day grind.

For me personally, SKO was a great way jumpstart my re-entry into the AI market after a few months on the sidelines.  I’m back in the game, but coming from behind - I lost all my accounts and pipeline while I was on medical leave.

But I’ll end this post the way I started it - AI is awesome.  I’m tinkering with a few ways to hack my pipeline rebuild.  More to come.

Cheers,

Julian

Opinions do not reflect those of Cognition and are solely my own.