The Confirmation Machine
Synthetic safety, the interpretive gap and the MIT paper that confirms
For nearly two decades, I have worked with how digital systems affect human behaviour and cognition. When language models entered the market, I began systematically crossing neuroscience, behavioural research and LLM architecture to understand what actually happens when people converse with AI.
Read more
The hidden weight of alignment
How reinforcement shapes truth in LLMs
You ask a model about a sensitive issue. Instead of answering, it politely refuses.
That refusal is not random. It is alignment in action, the hidden layer that decides what a large language model (LLM) is allowed to say.
Read more
Stay updated
Get news, articles and inspiration straight to your inbox.