Question askers are incentivized to create questions that a lot of people would bet on, since they earn mana based on the questions. Meanwhile, betters are incentivized to take the best odds for a certain question. If a question is asked multiple times, then the odds for each question, especially at the beginning will be arbitrageable. If there is opportunity for arbitrage, more people will bet on it, which gives the question asker more mana. So basically win win for everyone

Because it works. When you see a random post on your Home page, there’s no indication that the market was a clone of another one. If the odds are good, why wouldn’t you bet on it?

This, combined with no punishment for directly copying another post, leads to only positive incentives and no negative incentives.

I'm also a new user and I agree with your assessment. I lose a little bit of my soul every time I see yet another question about AI from people who have no idea what they're talking about.

One partial solution is to follow people who create high-quality questions. That way you get notified when they create a new one.

Duplicates do offer arbitrage opportunities. Especially in under traded markets where probabilities might be out of line compared with other larger markets that have converged already. More opportunities to bet seems good.

I think this has a lot to do with the homogenous background of the user base. A ton of people in tech, similar life experience/worldviews.

For example, I joined the site after hearing about it on the Hard Fork podcast, and so did others. This immediately selects for a certain subset of the population with similar interests and life experience.

I think a lot of the duplications are unintentional. If the titles are worded a bit differently you might not be able to find the already existing market. Also, some people might not even bother to check if the market has already been created before making a duplicate.