{"id":520,"date":"2026-01-25T05:42:21","date_gmt":"2026-01-25T05:42:21","guid":{"rendered":"https:\/\/squareright.com\/?p=520"},"modified":"2026-01-31T02:24:18","modified_gmt":"2026-01-31T02:24:18","slug":"ai-isnt-the-threat-monopoly-control-over-understanding-is","status":"publish","type":"post","link":"https:\/\/squareright.com\/?p=520","title":{"rendered":"AI Isn\u2019t the Threat. Monopoly Control Over Understanding Is."},"content":{"rendered":"\n

Artificial intelligence is being presented to the public as something almost alien \u2014 impossibly complex, autonomous, and far beyond the reach of ordinary people. The dominant narrative emphasizes fear, awe, and inevitability, while avoiding plain explanations.<\/p>\n\n\n\n

That framing serves a purpose.<\/p>\n\n\n\n

When people believe a tool is incomprehensible, they don\u2019t question who controls it. They don\u2019t experiment with it. And they don\u2019t challenge the systems built around it.<\/p>\n\n\n\n

AI is not replacing information systems. Data storage, data governance, and information architecture are not disappearing \u2014 they are becoming more valuable than ever. AI sits on top of these systems. It does not eliminate them. It amplifies power and control for those who already own and govern the information.<\/p>\n\n\n\n

AI is not the system itself \u2014 it is a tool that operates on existing information<\/strong>. Yet the public is repeatedly told that understanding it is far beyond their ability. What is rarely emphasized is that much of the foundational AI tooling is openly available, which is why competing language models exist and why companies around the world are developing their own systems.<\/p>\n\n\n\n

With modest resources and time, individuals and small teams can already experiment with these tools at a basic level. The barrier is not intelligence \u2014 it is the confidence to engage rather than defer. The confidence to tinker.
We used to be a society of tinkerers and builders. Monopoly control has trained us not to \u2014 encouraging us to consume and use tools on their terms, rather than understand how those tools work or recognize that experimentation is not an exclusive privilege.<\/p>\n\n\n\n

In the United States, experimentation with advanced technology is increasingly centralized \u2014 treated as something best left to large institutions and corporations. In contrast, some countries actively encourage widespread experimentation and technical literacy across their population.<\/p>\n\n\n\n

That difference has consequences. Innovation thrives when many people are allowed to tinker, question, and build. When experimentation is discouraged and control is concentrated, monopolies grow more powerful \u2014 and long-term competitiveness erodes.<\/p>\n\n\n\n

Fear of falling behind does not come from too much public understanding. It comes from systems that depend on limiting it to survive<\/strong>.<\/p>\n\n\n\n

\n\n\n\n

AI Is a Probability System<\/h2>\n\n\n\n

Before AI can be understood as a force shaping society, it has to be understood for what it actually is.<\/p>\n\n\n\n

At its core, modern AI \u2014 especially the kind most people interact with \u2014 is a probability system.<\/p>\n\n\n\n

Large Language Models do not \u201cthink\u201d in the human sense. They do not understand meaning, intent, or truth. What they do is calculate likelihoods.<\/p>\n\n\n\n

Language models work by breaking text into small units called tokens<\/em>. A token might be a word, part of a word, or even punctuation. Each token is represented numerically, and relationships between tokens are learned during training.<\/p>\n\n\n\n

When you type a prompt into an LLM, the system does not search for an answer. Instead, it calculates which token is most likely to come next, based on:<\/p>\n\n\n\n