Protecting the Thinking Process in Human-AI Interaction: The Forum Interstitiale Between Thought and Expression

International human rights law protects freedom of thought absolutely and freedom of expression subject to limitation. Doctrine calls the two sides of that line the forum internum and the forum externum. The line was drawn when thinking left no trace. It no longer holds. Within six months in 2026, a United States court ordered that twenty million ChatGPT conversations be produced under ordinary discovery rules, two federal courts reached opposite conclusions on whether a litigant's exchanges with a chatbot are protected at all, and the Supreme Court held that location records kept by a third party remain constitutionally protected. Records of where a person's body had been were protected. Records of how their mind worked were not. The law has no category for the search query, the exploratory prompt, the unfinished draft, or the confession made to a machine. The United States answer has been a quarrel about privilege: courts refuse protection because the machine is not a professional, like a doctor or a lawyer, and industry asks for a new privilege because it behaves like one. Both assume that protection depends on what the machine is. This article argues that it depends on what the person is doing. It proposes the forum interstitiale, a third zone between the two classical fora, protecting the thinking process under a standard stronger than ordinary expression law but short of absolute. Every candidate analogy in existing law is tested, from search queries and confidential dialogue to protected reading, aggregated records, the European law on correspondence and data protection, private diaries and drafts, and neural data. Each fails for the same reason. Protection attaches to a relationship, a disclosure, a recording, an addressee or a data type, and never to the function of the act as thinking. The zone only works upwards. It raises acts the binary abandons and takes nothing from the absolute core, which answers the objection that widening protection dilutes it. The article sets out who belongs in the zone, five elements of the protection standard, the duties those elements impose on providers, and a map of individual AI interactions across the three fora. This classification is not an academic exercise. It decides what can be demanded, kept, filtered, and used against the person who thought it. This working paper forms part of a personal research project on freedom of thought in the age of artificial intelligence. The public-facing companion to the argument is available at foruminternum.org. Comments are welcome at [email protected].