In an innovative experiment, a couple decided to outsource their Tunisia vacation planning entirely to AI, pitting ChatGPT against Claude in a head-to-head comparison to see which AI assistant could design the better travel itinerary. The couple established basic parameters—visiting Tunis, Tozeur, and Hammamet—then let each AI bot plan one full day in each city.
ChatGPT’s approach was cinematic and ambitious, starting with a sunrise walk through Sidi Bou Saïd complete with pastries and mint tea, followed by an intensive tour through the Medina and half-day exploration of Carthage ruins. In Tozeur, ChatGPT recommended visiting “Star Wars” filming locations in the desert and camel rides at sunset. The Hammamet portion focused on seaside relaxation, couscous feasts, and beach walks—creating what the author described as “ambitious, colorful, and a little chaotic.”
Claude took a more measured, culturally-focused approach. Its Tunis itinerary featured a guided food tour before a paced historical visit to Carthage. In Tozeur, Claude skipped the desert adventure entirely, instead recommending the Medina and a relaxed oasis afternoon. Claude also suggested two significant cultural stops the couple hadn’t considered: the Great Mosque of Kairouan and El Jem amphitheater. The Hammamet plan was stripped back to a Kasbah visit and early waterfront dinner.
When reality tested both itineraries, each AI showed strengths and weaknesses. ChatGPT’s ambitious Tozeur day proved exhausting, with a sandstorm blocking access to the “Star Wars” sets. However, its Sidi Bou Saïd recommendation delivered a picture-perfect experience. Claude’s food tour became a trip highlight, and its cultural detours to Kairouan and El Jem proved extraordinary.
The experiment revealed something unexpected: the AI bots reflected the couple’s own travel personalities. ChatGPT’s energetic, adventurous style mirrored one partner’s approach, while Claude’s thoughtful, practical planning matched the other’s preferences. Rather than declaring a winner, the couple concluded that the best journey combined both AI perspectives, creating a trip that accommodated two different travel rhythms and personalities.
Key Quotes
Recently, I’ve outsourced a lot of my life to ChatGPT — from work projects to personal errands. So, I thought, why not travel planning, too?
The author explains their motivation for the experiment, illustrating how AI assistants are becoming integrated into multiple aspects of daily life beyond traditional work applications.
ChatGPT’s energy mirrored mine: scattershot but adventurous, willing to gamble on exhaustion for the sake of a memory. Claude’s calm matched my partner: thoughtful, practical, deliberate.
This observation reveals how different AI models exhibit distinct ‘personalities’ in their recommendations, and how users may gravitate toward AI tools that reflect their own decision-making styles.
What they gave us was a way of traveling that reflected us back at ourselves — two personalities and two rhythms woven into one trip.
The author’s conclusion about the experiment, suggesting that AI tools don’t just provide information but can serve as mirrors for understanding our own preferences and decision-making patterns.
Our Take
This experiment provides valuable insights into the practical differences between leading AI models when applied to real-world scenarios. The contrast between ChatGPT’s maximalist approach and Claude’s conservative recommendations reflects broader philosophical differences in AI development—should AI push users toward novel experiences or prioritize feasibility and comfort?
What’s particularly fascinating is the unintended psychological dimension: the AI bots became proxies for the couple’s own travel philosophies. This raises important questions about AI as a decision-making tool. Are we choosing AI assistants that challenge our assumptions, or ones that validate our existing preferences?
The sandstorm incident also highlights AI’s current limitations—these models lack real-time data integration and cannot account for unpredictable variables like weather or road conditions. As AI travel planning evolves, incorporating live data feeds and risk assessment will be crucial for moving from suggestion engine to reliable planning partner.
Why This Matters
This real-world test of AI travel planning capabilities demonstrates how consumer AI tools are expanding beyond traditional use cases into lifestyle applications like vacation planning. The experiment reveals important insights about AI personality and design philosophy—ChatGPT’s more aggressive, experience-maximizing approach versus Claude’s conservative, culturally-sensitive recommendations reflect fundamental differences in how these models are trained and optimized.
The story highlights a crucial trend: AI assistants are becoming personalized lifestyle companions that users trust with increasingly complex, high-stakes decisions. The fact that a couple would entrust their vacation—a significant time and financial investment—to AI algorithms shows growing consumer confidence in these tools.
More significantly, the experiment demonstrates that AI tools may inadvertently reflect and reinforce user preferences, raising questions about how AI shapes our choices and whether these systems truly expand our horizons or simply mirror our existing biases. As AI becomes more integrated into daily life decisions, understanding these dynamics becomes critical for both developers and users.
Recommended Reading
For those interested in learning more about artificial intelligence, machine learning, and effective AI communication, here are some excellent resources:
Recommended Reading
Related Stories
- Sam Altman’s Bold AI Predictions: AGI, Jobs, and the Future by 2025
- Mistral AI Launches Le Chat Assistant for Consumers and Enterprise
- Elon Musk Drops Lawsuit Against ChatGPT Maker OpenAI, No Explanation
- AI Shopping Assistants Drive Surge in Online Holiday Sales
Source: https://www.businessinsider.com/used-chatgpt-plan-trip-my-partner-used-claude-2025-10