Eric Siu breaks down four AI proficiency levels for companies: unacceptable, capable, adaptive, and transformative. He argues businesses need to push teams to at least adaptive level, and shows how tools like ChatGPT and Claude can build agentic workflows that handle end-to-end tasks.
Here's the number one reason why businesses fail at AI. The major thing is 88% of companies rate themselves below capable when it comes to AI. They have no idea what to do. I'm going to show you what to do. So these are the 4 levels of AI proficiency at my company, Single Grain. If you look at the very bottom, you have people who are in the unacceptable level. That's the bottom level of the pyramid. It just means they're not adopting it at all. They're not using it. They're not even using like ChatGPT, not even using Claude. Capable would be someone who is, they're using some AI tools. Maybe they're using some Claude Skills. Maybe they're using Claude Cowork a little bit. You know, they're getting used to it. Then you have adaptive. These are people that are probably using Claude Code and they're able to build things on command. And then you have transformative. These are people who are building agentic, completely agentic end-to-end workflows. They might be using one of the Claws, for example, like an OpenClaw, right? I'm not saying you have to use any of these tools. I'm just giving you examples here. So the key thing is you have to make sure that you get people to at least adaptive because today's adaptive is tomorrow's capable. So get them to adaptive, if not transformative. If you want more tips on AI business and marketing, just comment newsletter.
