I asked ChatGPT and Claude 4 to plan my vacation to Tahiti. Here's how they compared.

In a unique exploration of AI capabilities, I turned to ChatGPT and Anthropic's Claude 4 for guidance on planning a vacation to the breathtaking islands of Tahiti. This holiday trip is a special celebration with my wife and a close-knit group of friends, honoring our dear friend Theresa's birthday. We will be enjoying a week-long stay in Moorea, surrounded by stunning landscapes and turquoise waters. To ensure a memorable experience, I sought advice on various activities for both daytime and evening, as well as recommendations for local dining and nightlife. Additionally, I asked for suggestions on how to celebrate Theresa's birthday in a special way. To validate the AI-generated suggestions, I enlisted Theresa and our friend Lisa to evaluate the responses from both chatbots. With my friends having already invested significant effort into planning this trip, their insights were invaluable in assessing the effectiveness of the AI suggestions. Theresa, the birthday celebrant, noted that both chatbots provided similar activity recommendations, including a cultural tour, 4x4 rentals, a lagoon cruise with snorkeling, and a sunset cruise on her birthday, which was a delightful surprise. ChatGPT suggested three restaurants: Rudy's, Moorea Beach Cafe, and the Manava Polynesian show, all of which we booked. Claude recommended Cocobeach, another spot we decided to try. While both chatbots mentioned Holy Steak House, the 40-minute taxi ride from our hotel made it less appealing compared to nearer options. When it came to presentation, I found ChatGPT's day-by-day itinerary format more user-friendly. In contrast, Claude's suggestions felt overly commercialized, focusing too much on the Cook's Bay hotel. Lisa also weighed in on the findings, praising ChatGPT for its thoroughness. With a sample itinerary that included pricing estimates and reference links, ChatGPT offered a more comprehensive view, albeit with some gimmicky icons. Claude's layout was visually appealing and easier to navigate, but it fell short in terms of variety, suggesting fewer activities and dining options. Interestingly, neither chatbot touched on the availability of scuba diving, despite its popularity in the Moorea region, where several party members planned to indulge in this activity. In the end, Lisa rated ChatGPT 4.5 out of 5 stars, while Claude received a score of 3.5, highlighting the strengths and weaknesses in their recommendations.