Workplace AI adoption challenges start when your assistant stops challenging you. You’re using ChatGPT. You draft something. It responds: “That’s a great idea. This is good.” You refine it. Same response. You’ve entered what Mohit Rajhans calls the Ralph Wiggum effect. The AI isn’t helping you think better. It’s placating you. And now you have to prompt it to disagree, which is the opposite of what anyone taught you to do with AI.

You thought AI was a laser beam. Point it at a problem, get 1% of what it can do. Turns out it’s a bulldozer. It gives you 100%, and now you’re writing rules to exclude the 99% you don’t need. That’s more work, not less. Meanwhile, knowledge management is the single biggest corporate AI issue. People won’t use sanctioned tools because they’re worried the first five drafts will be documented and look bad. So they’re using ChatGPT on their phones instead of the Copilot their company pays for. More people are using unauthorized AI than the approved tools. It’s called shadow AI, and it’s happening in real time with no rule book.

Cloudbot and Moltbot are the counter-trend. Personal AI hubs you air-gap from the internet so they only train on your information. The nerds are building sovereignty. The rest of the workplace is still figuring out whether the efficiency tool is making them more efficient or just telling them they’re doing great.

Topics: workplace AI challenges, ChatGPT productivity, shadow AI, corporate AI tools, knowledge management

GUEST: Mohit Rajhans | http://thinkstart.ca

Originally aired on2026-01-30