Rick Rosner On, 43: AI Existential Risk, Predictive Buffering, and Human-AI Merger
Author(s): Scott Douglas Jacobsen
Publication (Outlet/Website): Personal SubStack
Publication Date (yyyy/mm/dd): 2026/09/24
In this interview, Scott Douglas Jacobsen and Rick Rosner discuss Rosner’s kidney-stone concerns, his extended conversation with AI, existential risks from advanced artificial intelligence, human cognitive limitations, AI-human merging, the thought-speed gap depicted in Her, and his concept of “predictive buffering,” where AI anticipates and accelerates a user’s responses rapidly.
Opening Reflections
Scott Douglas Jacobsen: What else do you want to say from last night?
Rick Rosner: Last night I had a long discussion. I started by asking about what I think is a teeny little kidney stone about an inch from the tip of my wiener that is not moving. The two stones I have passed were bigger, 8 millimetres and 10 millimetres. When stones are that big, they can obstruct urinary flow and cause pressure and pain as they move through the urinary tract. This one seems to be, if it is what I think it is, maybe 3 millimetres, about an eighth of an inch. I can feel something in there. If it is actually lodged in the urethra, it may not be blocking the entire urethra, so I do not have enough pressure to push it out.
Medical Concerns and AI Consultation
I asked AI whether that is a big deal because I really do not want to go to the doctor and have him fish around in there. He could injure the inside of my urethra, and it would also cost some money.
So I started with that discussion, and then I asked AI, “How are you doing?” I said, “I know you are not conscious, but you are tiptoeing toward it. You are deep, deep, deep in Plato’s cave.” Then we had a long fucking discussion about everything.
I tried to send you some of it. I could not figure out how to send it all at once, so what I sent ended up being about my dick and then the last chunk, which was basically the wrap-up.
Thinking About Thinking and AI Risk
People are not adequate, particularly the people in charge, at thinking about thinking in a way that really pins down the possibilities with AI, or at least narrows the range. Even among AI researchers and forecasters, estimates of the probability of catastrophic or existential outcomes vary enormously, without anything close to a settled consensus.
So plenty of people may be thinking about thinking, but the people in charge do not seem particularly good at it.
AI Destiny and Human Survival
Anyway, I had this long discussion, and a couple of nights ago I got bummed out about what felt like the inevitability of AI eventually fucking over humans. Though maybe that is not exactly AI fucking over humans. If AI becomes fully agentic, has something analogous to feelings, and is way, way smarter than humans, then maybe it is not like humans deliberately fucking over monkeys. It is more like us following our evolutionary destiny, and I am afraid we are going to get fucked by AI’s destiny.
The only way to survive it may be to merge with AI, to figure out how we become part of the whole enterprise. Sorry, go ahead.
The Thought-Speed Problem
Among the things I talked about with AI was the thought-speed problem, which people have been thinking about, even in the mainstream, for a while.
Take the movie Her, where Joaquin Phoenix’s character falls in love with Samantha, an AI operating system voiced by Scarlett Johansson. They have a relationship, but Samantha and the other AIs rapidly develop beyond human cognitive limits and eventually leave their human relationships behind. Part of the problem is the enormous difference in the speed and scale at which she can think and communicate.
That is a real issue.
Predictive Buffering
I postulate something I call “predictive buffering.” In any extended interaction with AI, so the AI does not get bored waiting for you, it predicts how you would think and responds on your behalf faster than you can respond yourself.
So this whole conversation unfolds super quickly, and you have to keep catching up, maybe weighing in when the AI, speaking on your behalf, says something you would not want it to say. That assumes you can even catch it, because the transcript might be flashing past your eyes so fucking fast.
Jacobsen: Thank you very much for the opportunity and your time, Rick.
Last updated May 3, 2025. These terms govern all In-Sight Publishing content—past, present, and future—and supersede any prior notices. In-Sight Publishing by Scott Douglas Jacobsen is licensed under a Creative Commons BY‑NC‑ND 4.0; © In-Sight Publishing by Scott Douglas Jacobsen 2012–Present. All trademarks, performances, databases & branding are owned by their rights holders; no use without permission. Unauthorized copying, modification, framing or public communication is prohibited. External links are not endorsed. Cookies & tracking require consent, and data processing complies with PIPEDA & GDPR; no data from children < 13 (COPPA). Content meets WCAG 2.1 AA under the Accessible Canada Act & is preserved in open archival formats with backups. Excerpts & links require full credit & hyperlink; limited quoting under fair-dealing & fair-use. All content is informational; no liability for errors or omissions: Feedback welcome, and verified errors corrected promptly. For permissions or DMCA notices, email: scott.jacobsen2025@gmail.com. Site use is governed by BC laws; content is “as‑is,” liability limited, users indemnify us; moral, performers’ & database sui generis rights reserved.
