Machine Matthew L. Glossary
Machine Matthew L.
Machine Matthew L. is the tribute-act problem: an AI imitation of a particular human’s style, examples, judgments, and role performance.
The phrase is deliberately local. It names the fear more honestly than an abstract phrase like “AI replacement risk.” A model trained on lectures, comments, essays, prompts, jokes, grading patterns, course materials, and daily working habits might produce something that sounds enough like Matthew Langenkamp to be operationally tempting. It might answer student questions in a familiar register. It might draft memos with the right examples. It might even know that Taipei, auctions, San Miguel, China, and strategy cases are part of the voice.
But imitation is not the same as living contact with the world. A Machine Matthew L. can reproduce a corpus. It cannot go to the auction tomorrow, notice a new bidder, change its mind after a student question, sit with family difficulty, or be altered by the day unless a living human keeps feeding that loop.
The point is not paranoia. It is role clarity. If a system imitates a human’s visible role too well, the institution may forget that the person is not only a content generator. The person is an updating judgment system with accountability, embodied history, relationships, and consequences.
The answer is not secrecy. It is Anti-Replication Strategy: keep living, noticing, revising, and encountering the world. Do things the tribute act has not done.