Nathan M. Thornhill
Independent researcher and author. Former nursing assistant. Works on how you tell whether a mind is still there.
Fort Wayne, Indiana, United States
Actions
Nathan M. Thornhill is an independent researcher and author of Foundations of the Existence Threshold: The Scholarly Collection. He works without a university, a funding body, or a department.
His research addresses one question: how do you tell whether a mind is still there? He built a measure for it, then published the account of what it cannot do.
The question came from clinical work rather than philosophy. He spent years in healthcare, from nursing assistant to Long Term Care Administration and through ICU admissions, where families asked him whether their loved one was still in there. He never had a good answer for them.
The same question now applies to artificial intelligence, and that is where much of the work sits. He founded the Institute for Complexity Science and Advanced Computing (ICSAC), which publishes the research, and built CiteStamp, which catches the citations AI systems invent. He holds four provisional patents.
He is based in Fort Wayne, Indiana, and travels for speaking engagements.
Area of Expertise
Topics
Why your AI invents sources, and what actually catches it
Every team using AI to draft, research, or summarize has shipped a citation that looked authoritative and pointed at nothing. The journal is real. The authors are plausible. The DOI has the right shape. The paper does not exist.
This session explains the mechanism behind fabricated citations, why prompt engineering fails as a control for it, and what verification looks like when it is treated as an engineering problem rather than a wording problem. Worked examples throughout, including a live check against a citation graph that separates claims a human has signed from claims a machine inferred.
Attendees leave with a verification workflow they can apply the same week, a clear sense of which failure modes are fixable at the prompt and which require tooling, and the questions to ask any vendor claiming their product is grounded.
beginner to intermediate · 40 min
Do brains and language models decide the same way?
A rat commits to a choice. A macaque commits to a choice. A language model produces a token. Whether those are the same kind of event is an empirical question, and it is answerable.
This session walks through a cross-substrate comparison of decision dynamics: what has to be held constant for the comparison to mean anything, why the obvious way to run it produces a confound that looks like a finding, and what has to be fixed in advance to keep the analysis honest.
Attendees leave able to read cross-system claims about AI and cognition critically, including the specific confounds that make most such comparisons unreliable.
intermediate · 45 min
Still in there? Measuring whether a mind is present
Families in intensive care ask a question medicine still answers with a shrug: is my mother still in there? This session is about the attempt to replace that shrug with a measurement, and about what happened when the measurement was tested against its own limits.
The talk covers how a measure of integration and differentiation distinguishes conscious from unconscious states across a large body of clinical recordings, why it is not a consciousness detector despite being described as one, and what a pre-registered attempt to break it revealed about the information such measures discard. The same instrument runs on a language model, and the session closes on what that result does and does not mean.
Attendees leave with a clear picture of where consciousness measurement currently stands, what it can responsibly claim, and why the honest limits matter more than the headline accuracy.
general audience · 45 min
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top