I've spent the last month deep in AI models, workflows, and agent frameworks like OpenClaw and Hermes. As a business owner, I need to stay current with where the technology is heading, even if I have no interest in building AI-generated products myself.
My honest take: the tech isn't quite there yet. These systems still need to mature before they're reliable enough to trust with anything mission-critical. But that gap is closing fast, and it's only a matter of time before every business, especially small and medium-sized ones, runs on this stuff.
Which brings me to the conclusion I keep circling back to. In the near future, humans will focus on the two things AI struggles with most: taste and judgment.
I want to build physical products and brands that genuinely improve people's lives. But anyone running a small team knows the truth. You wear many hats, and most of those hats have nothing to do with the work that actually matters. The product, the design, the brand. The rest is admin, ops, follow-ups, and a hundred invisible tasks that quietly eat your week.
That's where these frameworks earn their keep. Not for the creative work, but for the boring, behind-the-scenes machinery that nobody sees and everybody has to do.
The shift this creates is bigger than productivity. Every founder will eventually compete on two skills: taste and judgment. The advantage of hard skills is collapsing. We're becoming curators instead of executors. Thinkers instead of doers.
For most of history, people who knew how to do things had the leverage. Now it's the opposite. The person who knows what to do beats the person who knows how to do it.
I'm probably making a video on this.
André