December 27, 2025
Five AI models scored 10 recent Gartner graphics. Clarity: 6.8. Promotional: 7.3. Likelihood of reversing the 80% failure rate: 2.7. I first asked today's question in 2011. Still waiting.
December 26, 2025
ConvergentIS just earned 9.7s and a perfect 10 from Spend Matters. But they've also walked away from major accounts that weren't ready. The scores prove the technology works. The exits prove that technology alone isn't enough.
December 26, 2025
Advantage comes less from bigger, faster tools and more from pre‑work that builds deep, agent‑level understanding before the snap.
December 25, 2025
Gartner's Guardian Agents: FOMO or Fact?
December 24, 2025
Gartner told 2.1 million followers that 80% of AI projects fail to deliver ROI. Then they projected spending will triple to $600 billion by 2029. They didn't connect the dots. I asked five AI models why that's annoying.
December 24, 2025
The 80% failure rate in enterprise AI isn't a technology problem. A new 65-page study from five elite universities just explained why—and what actually works.
December 22, 2025
Amazon Lens Live lets you shop what you see. In 1999, I used a ball camera to solve the same problem — but in enterprise procurement. The technology wasn't the solution. The SR&ED-funded theory behind it was. Agent-based thinking: 27 years waiting for non-human agents to catch up.
December 21, 2025
What is the difference between talking to AI and thinking with AI?
December 20, 2025
In 2008, I wrote that static systems fail because reality moves faster than they adapt. Very few engaged. This week, Snowflake invested $75-100 million in the same insight.
Why “Augmented Intelligence” Became the Invisible Third Rail of Enterprise AI
December 27, 2025
0
Human-in-the-loop models work regardless of technology era. Tech-led models fail regardless of technology era. That's not theory—that's 27 years of production data.