Jun 16, 2026 · 29m · we-live-to-build
He Built LLMs in 2016 and Still Trusts Gut Over AI
0:00 / 0:00
▶ Watch on YouTube →
gold bands on the timeline = statements, start to end. Hover to read, click to jump. CC turns on captions
In this podcast episode, host Sean Weisbrot and guest Robert explore the critical importance of human intuition in M&A due diligence, the operational limits of generative AI in marketing, and the broad societal implications of imminent technological disruption across traditional industries.
How this conversation actually went
Every chapter scored 0–10 on four independent dynamics. Hover any point for the reasoning behind the score. Sean holds 20.7% of the talking time here. How this is scored →
Sean as informed peer 3.1 Guest teaching 3.6 Guest disagreement 2.4 Sean pushing back 1.5
speaking balance: gold is Sean, purple is the guest (3 minute bins)
Sharpest disagreement ▶ 17:36 Robert dismantles the non-engineer private equity stance
Robert bluntly rejects the idea that any industry can avoid AI disruption, asserting that anyone claiming otherwise lacks engineering knowledge.
Hardest push from Sean ▶ 24:57 Sean reframes the Zerg archetype with Neuralink and Elon MuskSean challenges Robert's theoretical societal model by directly pointing out real-world corporate initiatives leading toward an invasive hive mind.
Biggest teaching moment ▶ 5:01 Robert clarifies real AI versus generative text hypeRobert educates the host on the distinction between true AI and basic text generation, referencing his team's work building language models as early as 2016.
Sean holds their own ▶ 9:30 Sean demonstrates technical learning from automated pitch testingSean details his hands-on experience building and tweaking a custom podcast workflow to prove where AI assistance ends and human personalization must begin.
the scores for every segment, with the reasoning behind each
| Chapter | Topic | Sean as informed peer | Guest teaching | Guest disagreement | Sean pushing back | Why |
|---|---|---|---|---|---|---|
| Uncovering Truth and Due Diligence in M&A | 4 | 3 | 1 | 1 | Sean asks about diligence challenges in M&A and contributes his own experience with startup investments failing due to hidden issues. Robert explains his structured 21-to-170 question framework and compares finding red flags to a massage therapist feeling knots. | |
| Risk Tolerance, Pivoting, and Capital Preservation | 2 | 3 | 1 | 0 | Sean prompts Robert on tolerance thresholds for failure in acquisitions. Robert outlines his playbook for pivoting companies during black swan events and preserving investor capital. | |
| AI Limitations in Due Diligence and Sales Outreach | 2 | 6 | 4 | 1 | Robert rejects the hype surrounding generative AI, noting that he was building LLMs back in 2016. He explains why AI cannot match proprietary agency copy data or spot deception during M&A deals. | |
| Sponsor Break: Network Before You Need It | 7 | 1 | 1 | 2 | Following a sponsor break, Sean explains in detail how he built his own podcast guest automation tool and diagnosed why automated pitches failed. He leverages this experience to ask how human relationship dynamics can be maintained in M&A. | |
| Designing AI and Robotics for Human Comfort | 1 | 4 | 2 | 0 | Robert delivers a monologue illustrating that human technology evolves around comfort rather than pure efficiency. He predicts household humanoid robots and cloned podcast avatars within years. | |
| The Universal Disruption of Traditional Industries | 3 | 7 | 6 | 2 | Sean introduces a private equity perspective that avoids AI-touchable businesses, which Robert forcefully dismisses as a non-engineer mindset. Robert insists that every industry—from accounting and law to construction—will be completely automated. | |
| Future Societal Trajectories and Starcraft Archetypes | 3 | 3 | 2 | 3 | Sean challenges Robert's optimistic view by highlighting that major corporate leaders exhibit psychopathic tendencies. Robert uses the three factions in Starcraft as an allegory for possible human technological futures. | |
| Neuralink, Flawed Leaders, and Dystopian AI Visions | 3 | 2 | 2 | 3 | Sean interjects that Neuralink and Elon Musk represent the path to a Zerg-like hive mind. Robert acknowledges leader fallibility and dystopian AI sci-fi before reflecting on practicing empathy as a deliberate skill. |
Statements from this episode (12)
Indries: Uncovering truth is the hardest part of M&A due diligence
“I think the hardest part is getting people to tell you the truth, because at the end of the day, you know, their goal is to get the highest multiple possible, you know, most bang for their effort in the last few years. So they're obviously biased. We are lucky…”
Indries: Due diligence framework utilizes nearly 170 total questions
“We have a standard of 21 questions, and, you know, we analyze stuff, and wherever we feel that something isn't adding up, we go deeper, and then we go deeper. Just, you know, for the sake of the audience, we have almost a 170 questions total, you know, that we…”
Indries: Firm Has Never Lost Client Capital on Any Deal
“So, so far we haven't lost our client's money. Not even once which is, you know, not easy, but we always try to at least make their money back and, or our money, you know, because we co-invest most of the time.”
Indries: AI Outreach Scripts Underperform Agency Scripts 90% of the Time
“90% of the time AI generated scripts underperform the scripts we have in our marketing agency.”
Indries: AI Cannot Detect Lies in M&A Due Diligence
“AI does not know to look through lies yet, right? So people tell you stuff, and you know it's a lie, but the AI will not know. It will not be able to figure out the lie.”
Indries: AI Capable of Deep Deal Analysis May Arrive by 2026-2027
“Maybe by the end of 20, 26, we'll have some AI, or maybe in 20, 27, but right now that's not the case yet.”
Indries: Indistinguishable digital AI avatars are one to two years away
“Give it maybe one to two years and digitally, I will be able to create a digital avatar of myself that looks exactly like me, sounds exactly like me. You will not be able to tell.”
Indries: AI will completely replace human accountants within 20 years
“Don't tell me that there'll be accountants 20 years from now. Why would you have an accountant when it's just math?”
Indries: AI will autonomously negotiate contracts and resolve lawsuits within 20 years
“You'll have, you know, AI just building the most amazing contract And the AI negotiating with other AI, you know, making compromises on each side, you know, just so that they can, you know, reach settlement or whatever. Right. When you will sue someone, 20 yea…”
Weisbrot: The leaders of the world's biggest companies have psychopathic tendencies
“The people that run the biggest companies in the world are, have psychopathic tendencies.”
Weisbrot: Neuralink Success Could Create a Zerg-Like Hive Mind
“I would say the Zerg would happen if Neuralink wins.”
Indries: People in Their Thirties Might Live for Hundreds of Years
“You and I might not die because we're like in our thirties. So we will evolve enough, you know, so that you can live to be a few hundred years old. And then if you live to be a few hundred years old, we might invent something, you know, that makes you basicall…”