The company the podcast is about and the cofounders run: a system that turns a spoken-word archive into a connected, cited site, with an LLM doing most of the work and a human reviewer keeping control of what is published. Episode 01 is its origin story, from one archive in late 2023 to the two problems, cost and trust, it was built to solve. The name comes from Mortimer Adler's Syntopicon. Start with episode 01 at about 16 minutes, where the two problems are set out.
Across the episodes
Enters as the series' own subject: an origin, a name, the two problems it was built to solve, and a first figure put on its value
Paul's first definition of the company names a stance rather than a product: the machine assists and judgment stays with people.
"and i think that shift to augmented intelligence versus artificial intelligence really is probably at the core of Syntopic."Paul Salazar, Syntopic, Episode 1 (11 Oct 2026), 5:46
The origin is custodial rather than entrepreneurial: Syntopic began as the problem of organizing one inherited archive, in late 2023, months after ChatGPT reached the public.
"And I found myself a few years ago in becoming the steward of someone's archive of thousands of hours of intellectual dialogues."Chad Nantais, Syntopic, Episode 1 (11 Oct 2026), 6:27
Frames everything that follows: the product is presented as answers to two problems, cost and then trust, not as a list of features.
"So this is a good opportunity for me to explain the two main problems that had to be solved here with Syntopic."Chad Nantais, Syntopic, Episode 1 (11 Oct 2026), 16:16
Paul's metaphor casts the product as restoring depth that compression stripped out, not as adding new content to a show.
"this is turning your MP3 back into vinyl."Paul Salazar, Syntopic, Episode 1 (11 Oct 2026), 20:06
The division of labor in one sentence: the model carries the bulk of the processing, and the structure around it keeps editorial control with a person.
"Syntopic uses AI, uses an LLM, to do some of the work, well most of the work, and having a harness so that the human still drives the process."Chad Nantais, Syntopic, Episode 1 (11 Oct 2026), 23:35
The review panel is the economic answer: a reviewer's attention goes only to what the earlier checks flag.
"and presenting those in a way with a quick, we built a quick review panel, so the human is only presented with what they need to actually check."Chad Nantais, Syntopic, Episode 1 (11 Oct 2026), 27:06
A second gate guards against harm rather than error, and where its line falls is the client's decision, not the system's.
"There's another one called the sensitivity filter that makes sure that nothing's sensitive and sensitive is defined by the client."Chad Nantais, Syntopic, Episode 1 (11 Oct 2026), 31:32
The episode's one quantified product claim, offered as a rough estimate that Chad explicitly declines to stand behind as exact.
"but just based on estimates, it looks like Syntopic creates basically eight times savings in money and time."Chad Nantais, Syntopic, Episode 1 (11 Oct 2026), 32:37
Presents the eightfold figure as a floor: the review step is the part of the pipeline expected to improve.
"It's only version one of the review panel."Chad Nantais, Syntopic, Episode 1 (11 Oct 2026), 33:18
Stakes out the company's scope: the frontier labs' race for authentic human text has a spoken-word counterpart, and that is the ground Syntopic takes.
"With Syntopic, we're focused on spoken word. So the same problem can happen with the spoken word,"Chad Nantais, Syntopic, Episode 1 (11 Oct 2026), 36:20
Connections
The name is an abbreviation of Adler's Syntopicon.
Chad tells the origin story.
Paul hosts and asks the questions.
Human review is the control layer the product is designed around.
Paul names provenance as one of the company's core values.
The McEwen–Dobbs dialogues were the first paying client's archive.