Most AI projects fail because people won’t use them, not because the tech doesn’t work. Scott Howard Swain builds AI around how humans naturally think instead.
The story
Here’s the thing nobody talks about: most AI integration projects fail not because the technology doesn’t work, but because nobody wants to use it.
We’re seeing a pattern. Companies drop serious money on AI tools, watch adoption rates flatline, then wonder what went wrong. The tech works fine in isolation. But when real people with real deadlines try to use it? Crickets.
Scott Howard Swain from Creative Robot figured out why. After 30 years in tech and running an empathy practice group with over 2,100 members across 650+ meetings, he noticed something: AI workflows fail when they fight against how people naturally think and communicate.
His approach flips the script. Instead of forcing teams to adapt to rigid technical requirements, he builds AI automations around natural human thinking patterns. Then – and this is the player-coach part – he personally mentors teams through adoption. No “black box” handoff. No crossing fingers and hoping it sticks.
The results are solid. Teams he works with often triple their efficiency within weeks. Not because the AI is magic, but because people actually want to use it.
His AgentFlow project makes this accessible – open-source custom instructions for deploying agentic AI development teams. And his EmpathyBot, trained since 2018, shows what happens when you combine cognitive empathy expertise with language models.
The trend we’re tracking: AI integration isn’t about finding the smartest technology. It’s about understanding the humans who need to use it. Companies that get this are pulling ahead. Those that don’t are burning budgets o
