Traffic is an asset, and some people are more interested in acquiring websites with a lot of traffic rather than MRR. But honestly, 300-500 is still too low, even if it is super targeted audience. That is the only assessment I can make based on the limited information you have provided.

Not every launch hits, but every try counts.

I build products in such a way that AI is JUST a layer on top of a product that can work independently of it.

But building ChatGPT wrappers is also a valid approach if you know what you are getting in to: chasing trends, aiming for a quick nice MRR, watching it all decline to zero, and then starting the cycle again.

That sounds exhausting to me, but in this case, the moat is not the AI, but rather the founders’ skills in building and shipping super fast and knowing which trends to ride.