[ghxst]

From my experience, there are tons of interesting places in my country that hardly get talked about online outside of my native language. You won’t really find them on the SEO-boosted travel blogs that google likes to return either. I feel like this is where AI could actually be useful.

[akdev1l]

If it’s not talked about much, it won’t be in the training dataset much

[ghxst]

From what I’ve seen, AI can do a solid job of bridging language gaps—though it depends on how it’s set up. The data exists, but it’s often buried in native-language sources. By focusing on local content, like newspapers and forum discussions you can shape a dataset that pulls in more authentic, region-specific data.

[cyberlurker]

But… isn’t the AI going to parrot those travel blogs?

[ghxst]

That's up to the engineers of said AI, if you goal is to have an AI that doesn't do that then you have multiple ways to achieve that goal.