<oembed><type>rich</type><version>1.0</version><title>VitorPamplona wrote</title><author_name>VitorPamplona (npub1gc…fnj5z)</author_name><author_url>https://yabu.me/npub1gcxzte5zlkncx26j68ez60fzkvtkm9e0vrwdcvsjakxf9mu9qewqlfnj5z</author_url><provider_name>njump</provider_name><provider_url>https://yabu.me</provider_url><html>You gotta give them guardrails. It&#39;s like AI in chess. Engines work because the game has a limited rule set and can explore possibilities ina closed space. When you are talking to an LLM you are *defining those rules of that &#34;game&#34;* for each session. You are literally making the game as you go. If the &#34;game&#34; is small enough, LLMs work. If it is too big or too undefined, there won&#39;t be enough tokens to do well.&#xA;nostr:nevent1qqsyeea64d0sdpmrjdf3xja8edv265lepxt73wguj4kpeslvss54k8spz9mhxue69uhkummnw3ezuamfdejj7q3q0mtatsat7ph6rsq0w8u8npt8d86x4jfr2nqjnvld2439q6f8ugqqxpqqqqqqz95fequ</html></oembed>