What Not to Use AI For: My Personal Recommendations

AI Fails: Because Even Super Smart Bots Can't Handle Your Drama (Or Your Ethics)

In a world where artificial intelligence is integrated into everything from drafting emails to planning trips, it’s tempting to lean on it for just about anything. But as someone who’s spent years working with AI and helping organizations implement it responsibly, I’ve learned there are definite limits. AI is a fantastic tool, but it’s not a substitute for human judgment in every scenario. Here are four things you’ll never catch me using AI for. These are areas where staying hands-on is crucial to avoid pitfalls.

1. Making Critical Decisions

AI is great for generating ideas or crunching data, but I never let it make the final call on high-stakes matters. Whether it’s approving a budget item, releasing a report, or charting a business strategy, the responsibility has to rest with me. AI can’t own the outcomes, especially when they involve money, reputation, or long-term impacts. My rule of thumb: The bigger the risk, the more I scrutinize. I use AI as a brainstorming aid, but I always apply my own expertise to ensure the decision aligns with reality.

2. Defining Values and Ethical Boundaries

AI operates on patterns from data, but it doesn’t inherently understand fairness, bias, or what’s right for my principles or my team’s. For instance, in hiring, an AI might flag a resume with a gap based on past trends linked to turnover. But I know that gap could represent valuable life experiences like caregiving, military service, or personal growth that add real diversity to a group. Handing over ethical calls to AI risks amplifying hidden biases. That’s why I keep human oversight front and center to make sure decisions reflect our true values.

3. Performing Reality Checks

AI can spit out answers with total confidence, but that doesn’t mean they’re accurate or up-to-date. I’ve seen it reference old tax laws or reversed court rulings as if they’re still valid. In areas like finance, legal advice, health, or policy, where facts evolve quickly, I always double-check against current sources. AI is a quick starter, but skipping the verification step because of its speed is a recipe for trouble. Treat it as a draft, not the definitive truth.

4. Navigating Human Relationships

AI might help refine phrasing or suggest responses, but it can’t grasp the nuances of empathy, history, or interpersonal dynamics. Building trust, fixing misunderstandings, or handling sensitive conversations requires reading the room. This is something AI just isn’t equipped for. Take a follow-up email after a project hiccup: AI could make it sound efficient, but it might miss the emotional tone if the other person is frustrated. In those cases, I pick up the phone or meet face-to-face. Authenticity is key in relationships, and AI can’t fake that.

I’m not saying we should use AI less. We should use it more thoughtfully. Before I turn to it, I ask myself: What if this backfires? Does it need my unique perspective? Am I building skills or just speeding through? Is it helping me think clearer, or am I offloading too much? And most importantly, can I stand by the result without blaming the tech? By drawing these lines, AI becomes a true partner rather than a crutch.

In the end, these boundaries help me leverage AI’s strengths while preserving what makes human input irreplaceable. What about you? Have you found spots where ditching AI led to better results?