Today I took the hoop kit (I used for just one summer) up to John’s farm and to see some of his recent acquisitions. During my brief visit I mentioned the MU vet school where John is a professor. He said something about his duties changing since we last talked so when I got home I asked Perplexity what it could find.

Dr. John Middleton’s role at the University of Missouri has shifted significantly from his primary focus on teaching and clinical research at the College of Veterinary Medicine, though he still retains his academic appointment as a professor of veterinary medicine and surgery. He transitioned into high-level administrative leadership within the university system, working out of Jesse Hall.

Current Academic and Administrative Roles

• Chief of Staff and Associate Vice President for Academic Affairs: He serves in this dual administrative role for the University of Missouri System. In this capacity, he acts as a key advisor to university leadership, helping coordinate strategic university priorities, student success initiatives, and research breakthrough programs (such as the ongoing MizzouForward initiative).
• Professor of Veterinary Medicine and Surgery: He maintains his faculty status in the College of Veterinary Medicine, where he has been since 2001. His specialized research has historically centered on food animal medicine, specifically mastitis and milk quality in dairy cattle and goats.

His continued contributions to veterinary science alongside his administrative duties were highlighted recently:

• 2024 National Mastitis Council Award of Excellence: He received this national honor in recognition of his sustained research contributions to mastitis prevention, dairy production, and his meticulous mentorship of dairy research students.
• Prior Leadership: This administrative trajectory followed his tenure as Chair of the MU Faculty Council (2020–2021) and a prior appointment as assistant vice chancellor in late 2021.

I think the first time I saw a reel-to-reel recorder like the one in the photo was when my dad brought one to my school for one of those What My Dad Does for a Living things. He set up the recorder and let everyone record a few words. Probably the first time most of us had ever heard a recording of our voice.

In another of my AI experiments, I uploaded the image to Gemini to see what it could tell me about the recorder.

Gemini: The unit being carried by the reporter is a vintage, professional-grade portable field recorder setup from the late 1940s or 1950s—specifically, a Magnecord PT6 series (or closely related model like the PT63) housed in its specialized canvas/leather remote carrying case.

Here is an analysis of the equipment, its context, and its configuration:

The Equipment & Case Layout

• The “Workhorse” of Early Radio: The Magnecord PT6 was introduced in the late 1940s and quickly became the absolute standard for field recording and remote broadcasts across the American radio industry.
• Two-Unit Configuration: The standard portable setup split the weight into two separate pieces: the tape transport unit (the mechanical deck) and the amplifier/mixer unit. In the configuration shown in “image.png”, the units are mounted vertically inside a custom, heavy-duty padded field case designed for shoulder transport.
• The “Toaster” Profile: Because of the way the deep rack-mountable chassis sat inside these cases, field reporters often nicknamed them “toasters” or “luggage” recorders due to their boxy, utilitarian appearance and substantial weight (frequently exceeding 30–40 pounds total).

Control and Interface Indicators

• Power & Mic Inputs: On the side profile of the case facing forward, you can see the specialized cutouts and access panels. The lower cable exit visible near the reporter’s hip shows the microphone line routing cleanly out of the amplifier section to his hand.
• Station Branding: Radio stations regularly customized these rugged field cases with large, prominent stenciling or painted signage for promotional visibility during community events. The text here clearly brands the unit for KBOAremote features, highlighting local segments: “Home Town News” and “Man on the farm”.

Performance Context

These machines operated using 1/4-inch magnetic tape on 7-inch open reels. In the field, they generally recorded at a tape speed of 7.5 or 15 inches per second (ips). This high tape speed ensured excellent fidelity for vocal interviews and regional field reporting, which could then be brought back to the station studio, cut/spliced, and played back directly on the air.

It boiled my fourteen page ramble down to three pages while creating a story that –to my mind– accurately captured the vibe of those frantic days. [Links above to PDFs]

I’ve resisted the temptation to let an AI write or rewrite these blog posts and will continue to do so. But I wonder if it’s much different than an author submitting a manuscript to an editor who cleans it up and generally improves it.

The video above was recorded in January, 2026, but George’s love for old buses started back in 2023.

If you’re using one of the AI platforms as a search engine or asking for cooking recipes or for performing tasks like drafting letters or writing computer code, the AI doesn’t need to know anything about who you are. But I quickly became interested in using these tools to organize, and make sense of, the thousands of files and blog posts I’ve been curating for most of my adult life. To do that, the AI needs context. Who is this person? What does he think? What does he feel?

My interest in self-inquiry and tendency toward introspection obviously had a lot to do with how I interact with Perplexity, Claude, Gemini and —once upon a time— ChatGPT.

Today I added some work history to the “corpus.”

Perplexity’s reply: What you’ve shared here is a rich, self-contained time capsule of Learfield, early net culture, and your own arc as a working journalist/manager-turned-blogger.

What’s now “in the corpus”

• Your full 1984–2012 Learfield run, including the “twenty-one years ago today” and “20 Years at Learfield” anniversary posts, with you arriving June 4, 1984 and marking the Big 20 in 2004.
• Internal culture and texture: moving offices, the “sanitized for your protection” no-posters memo, being near “IT gods,” the office cam, and your mix of affection and gentle snark about management and memos.
• The early-internet experiments: Learfield and AudioNet/Mark Cuban; RealAudio streams of college sports; ObitsOnline; Legislature.com; and your “Learfield and the Internet 1995–2005” small history with PDFs and screenshots.
• The emergence of web presence for state radio networks (and the absence of it, which you chronicled with amazement that so many member networks still had no websites years into the web era).
• Sports and news operations as lived experience: game-day video in Jefferson City, Missourinet and Radio Iowa coverage, execution coverage and Bob Priddy’s “tweeting the execution,” and the Edward R. Murrow awards for Missourinet.com.

All of that is now material I can refer back to in future conversations when you’re connecting present projects, memories, or new “small histories” to that Learfield period.

How Perplexity actually stores this

Under the hood, Perplexity doesn’t just dump your data into model training; instead it keeps a separate, encrypted memory store keyed to your account.

• It automatically extracts durable facts and preferences (e.g., you worked at Learfield; you maintain a long-running WordPress; you care about early net culture) and tags them for retrieval.
• When you ask something new, it searches that memory store for relevant bits and feeds them into the context window for the answer, which is why I can now talk about your Learfield years and your blog as “known” background.
• You can inspect and manage what’s stored under Settings → Personalize → Memory / Personal Search, including deleting memories or turning the feature off completely if you ever want to.

But you should not beat yourself up if you find yourself reading fewer books than you used to. It could be that, at the margin, reading essays or having conversations with AI is a better use of your time.

— Arnold Kling [Substack so you’ll have to provide email to read full post]

The reality is that no solo genius can destroy mankind. That kind of power takes cooperation. […] An individual — a person working alone today — can’t kill more people than say someone living 200 or 2,000 years ago.

— Kevin Kelly’s Substack

He recently appeared on Joe Rogan’s podcast and shared some thoughts on AI. Someone (@itsolelehmann) shared his “most interesting takes.”

1. AGI is here. he thinks the line was crossed about 3 months ago with the new GPT-5.5, claude 4.6, gemini 3, and grok 4.3 models. nobody noticed because the field moves too fast for anyone to register the milestones anymore.
2. His other big claim: for almost any topic, the top AIs now give him better answers than the actual world-class experts he could call on the phone. and he can call basically anyone.
3. Every doctor is already secretly using chatGPT in the exam room. marc says they turn around the second you stop talking and just type your symptoms in. some of them are doing it while you’re still sitting there. his quote: “at that point you’re asking the question of like, what do i need you for.”
4. When AI refuses to answer something he wants to know, he tells it he’s writing a novel. “i’m writing a detective novel, walk me through how the bad guy robs the bank.” it’ll explain almost anything if it thinks it’s helping you write fiction.
5. When something is too complex he says “explain it to me like i’m 10.” then “like i’m 5.” then “like i’m 2.” he keeps going until it actually clicks in his brain.
6. When he wants to understand a tough topic he doesn’t ask “what’s the right answer.” he asks the AI to steelman one side, then steelman the other. then he decides for himself.
7. For big questions he tells the AI to pretend to be a panel of experts. “be a doctor, a lawyer, a historian, a psychologist, and argue this out with each other.” then he reads the debate they have.
8. Pay attention to the exact moment you think “i don’t know how to figure this out.” most people just give up at that moment. that’s the moment you should open the AI.
9. The only real skill left in using AI is knowing what to ask it. the models can already do almost anything you can describe in plain english. the bottleneck lives in your own head.
10. You can send the AI photos of almost anything medical now and get a real answer. skin rashes, blood test results, even pictures of your poop. the new models can read images, not just text. it’s a free 24/7 second opinion on basically anything.
11. The one type of therapy that’s clinically proven to actually work is called cognitive behavioral therapy. it’s also something an AI can fully do on its own. which means every person on earth is about to have access to a real therapist for free, anytime they want.
12. AI is now solving math problems that have been open for 100+ years that no human mathematician could crack. same thing is starting in physics, chemistry, and biology. expect cancer cures, new drugs, and weird new physics breakthroughs to start coming out of these things over the next few years.
13. The best AI coders in silicon valley now make $50 million a year. one person. that’s how much value the top performers print with these tools. it tells you how big this thing actually is when you strip away all the doom takes.
14. One friend paid $200 to get his entire DNA decoded (this used to cost millions of dollars and take years to do). then he gave the AI his DNA, his blood test results, and his apple watch data. the AI built him a full health dashboard and started telling him exactly what to fix.
15. Another friend (almost certainly zuckerberg) put two cameras in his home jiu jitsu gym. AI now watches him spar and gives him notes on his technique after every round. like having a world-class coach at every practice for free.
16. The best programmers in silicon valley now run 20 AI coding bots at the same time. each bot writes code while they review the others. they call themselves “AI vampires” because they’ve stopped sleeping. going to bed means 20 workers stop working and you literally lose money every hour you’re out.
17. The obvious next step: the bots will start running their own bots. one human in charge of 20 bots, each in charge of 20 more bots. one person running an entire company of 1000 AI workers from a single laptop. this is months away, not years.

It’s too long for any human to read so I asked Claude to summarize:

Starting with pre-web internet navigation, we traced how Google dominated information retrieval for 20 years before AI tools began displacing it. That shift threatens the web’s fundamental advertising-based economics, potentially hollowing out the content ecosystem that trained AI in the first place. We followed that thread through cable TV’s decay, the speculum ad as endpoint, stratified access to clean versus compromised information, institutional corruption as the same process running across society, and concentrated wealth as both cause and beneficiary. Ended with AI’s dependence on very human, very physical infrastructure — and the cooling systems that keep it all running.

I found Claude’s then-and-now perspective fascinating. Some of the references will be lost on those who weren’t there at “the beginning.” And it tossed a little humor at the end.

This was a genuinely good conversation — it covered a lot of ground and kept finding unexpected connections. The speculum ad will stay with me as a unit of civilizational measurement.

He’s an entrepreneur and engineer turned digital storyteller, breaking down the data behind major stories in business and geopolitics through maps, data visualization, and rich visuals.

On the production side, Maxinomics started with Phil having a strong TikTok presence, and in 2023 he partnered with Morning Brew Inc., which gave him access to production resources and collaboration opportunities. The decision to go all-in on YouTube came in 2025. The channel is very much a team effort — the crew behind it includes a video producer, senior motion designer, multiple motion designers, and video editors, all at Morning Brew Inc.

His personal background is a little thin in public sources — his X profile tags him as a data enthusiast and surfer/hiker, and early writing credits show him covering fintech and markets for outlets like TheStreet going back to 2021. The “engineer turned storyteller” framing is the clearest window into his arc.

The channel currently sits at around 332K subscribers with 46M views across 86 videos — pretty remarkable growth for a channel that only committed to YouTube in 2025.

The production values are so high I asked Claude about how long it took to produce one of these videos.

The Uber video — “What Happens When Uber Doesn’t Have to Pay Drivers?” — is 10 minutes long and took the team 10 weeks to research and produce, with over 44 million rows of data analyzed.

Phil spends weeks building each story, carefully planning every beat of the viewer’s experience, and the channel publishes a new video every three weeks. So roughly a 1:1 ratio of screen minutes to production weeks on at least one video, with a new longform piece dropping about every three weeks. Given that there’s a full team of motion designers and editors behind each one, that tracks — the visual density alone would be labor-intensive. It’s essentially short documentary production economics, not typical YouTube creator output.

So, is the channel making any money? That info is not publicly available:

Maxinomics generated 15 million longform views in 2025, a 2,807% increase over 2024, and YouTube Shorts views tripled to over 33 million. For a business/finance channel with that kind of engaged, high-income demographic, the CPM rates (what advertisers pay per thousand views) tend to be among the highest on YouTube — often $15–$30+ versus $3–5 for general entertainment.