<oembed><type>rich</type><version>1.0</version><title>Scott Jehl wrote</title><author_name>Scott Jehl (npub133…6rqrd)</author_name><author_url>https://yabu.me/npub133e0khgx56jygttzm705s8l377g7j2zcnuj8e7yn92h59c9keheqe6rqrd</author_url><provider_name>njump</provider_name><provider_url>https://yabu.me</provider_url><html>Like, yes, they make stuff up sometimes. They also can for example, in 10 literal seconds, list 20+ family-friendly activities happening in NYC on a given saturday–a task that would have taken an hour sifting through bookmarks and searches. (Swap in a thousand other use cases.) Just seems like the output is the smallest problem among a lot of very big ones that make me not want to use LLMs, yet it gets so much attention?</html></oembed>