Founder of MASONS.ai. Writing on agency, networks, and what happens to the economy when software stops waiting to be told what to do.
I know by the time this is in front of you, OpenClaw has already gone from the thing everyone was talking about to the thing everyone is moving past. I've been observing and monitoring until the dust settled, and here are my thoughts on OpenClaw.
You've probably heard about "Next-Token Prediction", the mechanism behind all the Large Language Models (LLMs). By predicting the next token, generative models create coherent text. It disrupted discriminative models, which were limited to recognizing patterns within predefined boundaries.
The latest advancement in AI models centers on reasoning capabilities, embodied in Chain of Thought (CoT). Through hashtag#CoT, models break down complex problems into logical steps, making their reasoning process explicit, which aids in analysis and debugging. This technique has improved the performance of AI models as of early 2025, notably GPT-o1, GPT-o3, and DeepSeek.