We used several AI search engines, including Gemini, a large language model that can handle files of up to 750,000 words, and Semantra, an open source “semantic search engine” developed by Mr. Freedman. Instead of looking for specific terms—“sustainability,” say, or “climate change”—it looks for concepts or themes. “It’s a new paradigm of search, which doesn’t look at keywords, but tries to capture meaning,” said Mr. Freedman.
The Times has specific policies on the use of artificial intelligence, and nothing that comes directly from an AI program can appear in our articles, in part because of the possibility of hallucinations — pretty much the program just makes things up. So after running our queries through these search engines, Mr. Seward and Mr Freedman delivered the results to Ms Mzezewa. From her point of view, the technology has been very helpful in spotting interesting nuggets within this mountain of text, such as the impact of world events on the list, as seen in the 2009 inclusion of Kabul, the capital of Afghanistan (we called it a “fragile city on the road to recovery”).
She found Semantra especially useful “because it gave more context over time,” she said, and let her see how we’d written about topics like hypertourism and the rise of social media in travel, even if we hadn’t used those exactly words.
For example, we asked AI programs to identify instances where we had written about sustainable travel. This term didn’t really exist when the list started, but the concept of greener travel did. Among the examples found by the search engine was Star Island in the Bahamas, which first made our list as ‘eco-destination of the year’ in 2009.
When we humans started looking back over the years of the lists, a few themes jumped out: The impact of smartphones and social media, the growing focus on climate change, and the potential negative impact of travel, including overpopulation. The analysis of AI programs largely mirrors ours, providing a kind of high-tech support to our reporter’s intuition.
Selecting our list each year is a team effort that requires knowledge of travel trends, an eye for great graphics, and a sense of what people are looking for in their travels now — to name just a few of the skills they’ve acquired. AI won’t choose the Places you’ll go anytime soon, but it can help us understand where we’ve been.
