Fun to have Dots to create games while working on other things.

AI can give you a very convincing answer. Full explanation, neat bullet points, even links. Still worth asking: does the source actually say that? 😅 Try asking it to separate: • what the sources support • what it’s inferring • what it couldn’t verify Then open the links yourself. Asking AI to check itself is useful, but it can repeat the same mistake. The diagram below is a simple way to think about it.

In love with the current @NousResearch desktop. So fun to work with in such UI. Was considering Grok bot but Hermes is more customizable for me.

First try using Codex + Higgsfield MCP. Not bad 😅

Every @openclaw update is a make or break sometimes 😅

Cool

Openclaw orchestrate > Codex build > Claude Code review > Openclaw verify > Repeat. Trying it first time here. Let's see if things can be done once I wake up 😂

Trying out Cowork dispatch feature. Quite cool and convenient especially for lazy people like me. 😅

We've spent years talking about AI changing the world. GPT-5.4 scoring above human baseline isn't a headline. It's a checkpoint. The question isn't whether AI will change how crypto works. It already is. The question is whether you're using it yet.

😂