Artificial intelligence is now advising people on how to spend their free time, and the results are… mixed. A recent experiment testing three AI systems – Claude AI, Google Gemini, and ChatGPT – on hobby recommendations revealed a surprisingly practical, yet occasionally bizarre, range of suggestions. The prompt was simple: a 39-year-old Los Angeles resident with typical interests (travel, gaming, reading) seeks a new pastime, mindful of a 9-to-5 job and a knack for finding good deals.
The AI Consensus: Gardening and Reselling
All three AI models converged on gardening as a low-effort, low-cost option. This is notable because it aligns with broader trends in suburban leisure: gardening has seen a surge in popularity, particularly among those seeking cost-effective hobbies. The algorithms also recommended reselling vintage items, a pastime that taps into the growing thrifting and resale economy. Gemini further refined this by suggesting pairing it with travel rewards hacking – a logical connection given the user’s stated interest in deals.
Divergent Approaches: Gemini, Claude, and ChatGPT
Claude AI stood out for its targeted approach. It directly appealed to the user’s deal-seeking personality, framing gardening as a way to save money. This demonstrates AI’s increasing ability to tailor suggestions based on inferred behavioral traits. Gemini took a holistic approach, suggesting hobbies that complement existing interests, like beer brewing alongside gardening.
ChatGPT, however, strayed into the absurd. It recommended amateur astronomy in light-polluted Los Angeles and, more bafflingly, beekeeping. The latter suggestion is particularly questionable; while beekeeping has gained traction in some urban areas, the time commitment (often exceeding the AI’s claim of 2-4 hours monthly) and regulatory hurdles were glossed over. This highlights a key limitation: AI can generate plausible-sounding suggestions without deep contextual awareness.
The Problem with Algorithmic Leisure
The experiment underscores a larger trend. As AI becomes more integrated into daily life, it will inevitably shape how people spend their time. The issue is not whether AI can suggest hobbies, but whether those suggestions are truly aligned with individual preferences and realistic constraints. The disconnect between the AI’s logic (e.g., assuming gamers moonlight as beekeepers) and real-world feasibility raises questions about the quality of automated advice.
Ultimately, while AI can surface ideas, the human element – informed by actual experience and nuanced judgment – remains essential. The algorithms can point you toward a hobby, but they can’t tell you whether it’ll actually stick.
