You're signed in. Welcome to kevinsomany.com.
Explanatory articles on frontier AI — each one defines every term the moment it lands.
An agent is not simply 'an agent.' What matters is how agentic it is: how much freedom it has to decide what to do next. Here is the spectrum, from fixed business workflows to humanoid robots, and why more autonomy always needs more control.
Read article →Before an AI model is connected to any tool or document, it runs on pre-trained knowledge: patterns compressed from an enormous pile of public text. Here is what that background knowledge does well, where it fails, and how grounding fixes it.
Read article →Much of today's AI boom comes from machine learning: software that improves by studying examples rather than following hand-written rules. Here is what that means, how it powers everything from spam filters to chatbots, and where it actually creates value.
Read article →A normal chatbot gives you an answer. An agentic AI system works through a task — planning, using tools, drafting, reviewing, and revising before it hands anything back. Here is what that means, and why the second approach works so much better on hard problems.
Read article →The gap between AI novices and power users is not the tool they pay for. It is whether they hand the model a real brief, and whether they let it argue back instead of just flattering them.
Read article →The numbers on AI are staggering, and easy to get wrong. Here is what the primary research actually says about its economic impact, and how to tell scaled value from hype.
Read article →