Standing in the chaos of our new Manhattan apartment, surrounded by boxes and staring at a broken dishwasher, the things I desperately needed were brutally simple: scissors, unpacking space, and a plumber. ChatGPT hadn’t even registered on the list of priorities.
Our family had just completed a whirlwind spring break move into a significantly smaller home, and the experience delivered a powerful lesson about the limits of artificial intelligence. It highlighted a gap between what AI *can* do and what truly matters when facing real-world challenges.
We’d previously used AI tools like Claude to analyze purchase agreements and Gemini to visualize our daughter’s new blue bedroom. But those digital assists vanished in importance the moment we began wrestling with actual possessions and the realities of limited space. A neatly formatted, emoji-filled furniture plan from ChatGPT was quickly abandoned.
Ironically, our disconnection from AI was partly intentional. The day before the move, I shut down all our computers, including the network hosting my personal AI assistant. As my digital world went dark, I found myself unexpectedly thrust into an “AI fast.”
Even with the systems online, the sheer logistics of the move – coordinating with movers and building management – consumed every available moment. There simply wasn’t time to formulate prompts for ChatGPT or Claude on a mobile app.
Accepting that AI wouldn’t be much help, I leaned into the analog world. It felt surprisingly liberating.
Our first day in the new apartment included a trip to Bloomingdale’s for bedding. We didn’t upload photos of duvet covers to AI for comparison. Instead, we sought the expertise of Sylvia, a knowledgeable sales associate who guided us to the perfect choice – Matouk, for those familiar with quality bedding.
For interior design, I bypassed AI-powered room planners. We scheduled a consultation with Elena, a professional who visited our home, meticulously measured the space, and offered practical advice. Unlike AI’s tendency to ignore physical constraints, Elena wasn’t afraid to say “no,” gently suggesting our dining room chairs wouldn’t work in the new space – and she was right.
When the dishwasher sprung a leak, I didn’t turn to Claude for a diagnosis. Our building superintendent connected us with Igor, a cheerful plumber who quickly identified and fixed the problem. He even tightened a loose faucet and repaired a faulty burner, refusing payment for the extra work.
The week was exhausting, but profoundly satisfying. Solving problems with experienced, helpful people felt genuinely good. They offered valuable pushback when needed, and delivered tangible results that stood up to scrutiny – something AI couldn’t replicate.
This isn’t a rejection of AI. I continue to use it daily for tasks like email triage and organizing thoughts, and rely on it for technical troubleshooting. It remains a valuable tool in many areas.
However, for tasks requiring practical skill, nuanced judgment, or simply a human touch, I’m firmly choosing people. The experience reinforced the enduring value of human expertise and connection.
