Ken Tannenbaum AI Literacy for the Workplace

The most dangerous way to use AI at work is the most natural one: to trust it. These tools are built to be trusted. They speak fluently and confidently, in complete, well-organized sentences - like a knowledgeable colleague who's done the reading. And most of the time, they're genuinely useful, which is exactly what makes the trap so well-disguised: a tool that's right most of the time teaches you to stop checking, and the moment you stop checking is the moment it produces something wrong with the same fluent confidence it produced everything right. The lawyer who filed a brief full of citations his AI invented didn't distrust his tool too little in general. He distrusted it too little in the one instance that counted. AI Literacy for the Workplace is about the judgment that prevents that - the practical skill of using AI well and safely. It is not a technical book; you won't learn to build a model, and you don't need to. You'll learn how these tools actually work well enough to judge them, where they characteristically fail, how to verify what they produce, how to navigate the new risks they bring into the workplace, and above all when to trust them and when not. We're past the novelty phase and past the productivity phase. We're in the consequence phase - the output goes straight into the world, the failures arrive with names attached, and the law has arrived too. Under the EU AI Act's Article 4 and NIST guidance, AI literacy is now a workplace requirement, not a nice-to-have. Inside, you'll learn how to: Understand why models sound so confident, why they hallucinate, and how training data, cutoffs, and bias shape what they sayBuild the core discipline - never trust, always verify - into concrete habits for sourcing, catching fabrication, and keeping a human in the loopNavigate the new threat surface: deepfakes and synthetic media, AI-powered misinformation, prompt injection, and the data you must never paste into these toolsDevelop the judgment for when to use AI and when not to - how stakes change the rules, when to disclose, and who owns the outputExtend literacy across an organization through policy, training, and role-by-role knowledge - and meet what the EU AI Act and NIST now expect This is not a tool tutorial, not a prompt-engineering manual, and not an anti-AI book. It's written by someone who uses these tools every day. The argument isn't that AI is bad - it's that AI without judgment is dangerous, which is a very different claim. Who it's for: every working professional who uses AI tools (or whose tools have quietly started using AI on their behalf); managers and team leads responsible for how their people use them; and anyone responsible for AI policy, compliance, or training - in HR, legal, operations, or IT - who now has to ensure AI literacy across an organization, often because the law requires it. The value of AI at work is real, the danger is real, and the only thing that separates the two is human judgment that stays in the loop. Accurate as of mid-2026, with the still-moving legal specifics flagged. Part of the Finnoybu Press AI for Everyone series. The lawyer had everything except the judgment. Let's make sure you don't share his fate.