Feb 19, 2026 · 40m · no-priors
The AI Code Slop: Risk or Opportunity?
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 episode of No Priors, hosts Sarah Guo and Elad Gil dismantle the 'SaaS-pocalypse' narrative, analyzing the operational challenges of AI code generation, shifting engineer identities, and the durable moats required for enterprise software defensibility. They also explore the macroeconomic impact of unprecedented revenue velocity across frontier labs, collapsing inference costs, and historical parallels from past tech cycles.
How this conversation actually went
Every chapter scored 0–10 on four independent dynamics. Hover any point for the reasoning behind the score. The hosts hold 99.9% of the talking time here. How this is scored →
The hosts as informed peer 8.3 Guest teaching 2.9 Guest disagreement 1.8 The hosts pushing back 2.2
speaking balance: gold is the hosts, purple is the guest (3 minute bins)
Sharpest disagreement ▶ 28:08 Challenging the long-tail thesis with power law distribution
Elad directly interrupts and rejects Sarah's claim that the long tail will dominate market cap creation, citing historical ad revenue and startup power law concentration.
Hardest push from the hosts ▶ 11:07 Refusing the premise of autonomous agent software purchasingElad strongly pushes back against recent market hype claiming autonomous agents are purchasing enterprise software, reclassifying the behavior as standard business development deals.
Biggest teaching moment ▶ 15:26 Empirical breakdown of revenue velocity and token cost collapseElad provides detailed historical benchmarks showing AI labs hitting 10B revenue in roughly one year versus decades for legacy tech, alongside 88x to 150x token price reductions.
The host holds their own ▶ 33:52 Historical analysis of Lotus 1-2-3 displacement and defensive bundlingElad demonstrates deep historical software domain expertise by drawing direct parallels between Lotus's sudden eclipse by Microsoft Excel and current AI startup vulnerabilities.
the scores for every segment, with the reasoning behind each
| Chapter | Topic | The hosts as informed peer | Guest teaching | Guest disagreement | The hosts pushing back | Why |
|---|---|---|---|---|---|---|
| Debunking the Myth of Vibe Coding and Vibe Sales | 8 | 2 | 1 | 2 | Sarah and Elad examine the 'SaaS-pocalypse' panic, agreeing that replacing complex vertical enterprise workflows with vibe coding and vibe sales is deeply overstated in the near term. | |
| Startup Reality vs. Enterprise Complexity in Software | 8 | 2 | 2 | 2 | The hosts explain why small 5-person startup coding habits do not extrapolate to Fortune 100 enterprises due to security, change management, and governance overhead. | |
| Infinite Software Demand and Evolving Engineer Identities | 8 | 3 | 1 | 2 | Elad details how software demand absorbs increased engineering productivity, while analyzing how changing workflows will alienate bespoke craftsmanship-oriented engineers. | |
| Agent-First Engineering and Code Quality Governance | 8 | 3 | 3 | 3 | Sarah outlines production code fragility from AI slop, while Elad pushes back on exaggerated claims that autonomous agents are already executing enterprise purchasing decisions. | |
| Historic Revenue Velocity and Collapsing Token Economics | 9 | 4 | 1 | 1 | Elad shares empirical data showing historic revenue acceleration to 10B and 100B alongside staggering 150x and 88x collapses in token pricing for frontier-level models. | |
| Exponential Inference Scale and Biological Energy Efficiency | 8 | 3 | 1 | 2 | The conversation compares data center inference energy draw against the human brain's 12-20 watt efficiency and projects tech expanding to 20-30% of global GDP. | |
| Platform Forward Integration and the Power Law Debate | 8 | 4 | 4 | 4 | Elad challenges Sarah's premise about the long tail dominating AI software market cap, arguing that power law concentration and platform forward integration will eat vertical apps. | |
| Lessons from the Dot-Com Crash and Strategic Exit Planning | 9 | 3 | 1 | 2 | Elad extracts strategic exit lessons from the dot-com era, advising founders to institute regular, emotion-free board discussions to navigate peak valuation windows before displacement. | |
| Historical Cycles, Product Bundling, and Moat Defense | 9 | 3 | 2 | 2 | Elad cites the rise and fall of Lotus 1-2-3 against Microsoft Excel to argue that multi-product bundles represent the most potent defense against platform model provider commoditization. | |
| Compressed Innovation Velocity and Modern Control Points | 8 | 2 | 2 | 2 | The hosts conclude by dismantling the SaaS dogma of 'do one thing well' in a world where two years represents a decade of technological displacement, emphasizing true control points. |
Statements from this episode (21)
Gil: Decagon and Sierra Represent a Shift Toward Utilization-Based Support Agents
“Ultimately, I think Decagon and Sierra are examples of companies where you're moving from proceed software to basically utilization based customer support related agents, right? That is a real shift. That may impact some of the prior wave of sort of perceived …”
Guo: Portfolio Firm Scaled to ~100 Salespeople with Under 50 Engineers
“I, we have portfolio companies with, you know, hundreds of millions of dollars of revenue who are very committed to as much token usage as we can, As few great people as we can have, and today, you know, we have less than 50 engineers, and they went from zero …”
Gil: 'Vibe Sales' Is Not Happening Anytime Soon
“Oh yeah, Vibe Sales is definitely never, it's not happening anytime soon.”
Gil: Fortune 100s will probably not replace CRMs with vibe-coded software
“Does that mean that a fortune 100 company is going to displace their CRM with some internal thing that got Vibe coded over the weekend? Probably not. And so I think it's also Extrapolating or projecting behavior of very small technical startups onto the world'…”
Guo: Eliminating code bottlenecks overstates the threat to enterprise software vendors
“I actually think the idea that actual production of code becomes not the bottleneck for if you know what the spec is, not the bottleneck is like incredibly interesting, but I do think it overstates like how much of the overall software vendor problem that is.”
Gil: AI engineering productivity boosts will be absorbed by infinite software demand
“I think that fundamentally you know, there's so much demand for software and there's so little supply of engineering and reality relative to that demand that as you add this enormous boost of productivity to software engineers it just gets soaked up. Right. Be…”
Gil: Craftsmanship-focused engineers will become unhappy at large companies due to AI
“I think we're going to see a lot of these really great engineers who care about the bespoke craftsmanship of everything they do. They're going to be unhappy working at larger companies. As these coding tools get even more. Accelerant because it goes against th…”
Guo: High-status engineering tasks can be surprisingly easy for AI agents
“One thing I've seen is that if you have An engineering identity that's based on that, like a value based ranking of difficulty or skill, like the specific types of engineering that are considered, you know, impressive or high status can actually be like less h…”
Guo: Unreviewed AI-generated code introduces fragility into production codebases
“If you can generate an enormous amount of code and no one is reading it, you don't know the quality of the code. Nobody deeply understands the code base and there's more fragility, right? It's like the slop problem, but instead of it being like vibe coding slo…”
Gil: Claims that AI agents make vendor purchasing decisions are overstated
“I think the one other thing people are bringing up that I don't quite buy is that agents are already making like big decisions for vendor purchases and things like that.”
Gil: AI labs scaled from $1B to $10B revenue in one year
“ADP Took 20 something years to grow from a billion to ten billion in revenue. And then the next wave of companies like Adobe took about 20 years to go from one to 10. And then you fast forward in time and you have things like Salesforce or SAPs for an even mor…”
Gil: GPT-4-level token pricing dropped 150x in 21 months
“We looked at the cost of a GPT-IV level or equivalent model. We looked at that a year or two ago and basically in 21 months, it went from like 37 bucks for a million tokens to 25 cents. And so, you know, pricing dropped by a 150 X in 21 months.”
Gil: o1-equivalent model inference costs dropped 88x in 11 months
“The cost of a million tokens on an O-one equivalent model in December of 24 was about 26 bucks. And then in November of 25, it was 30 cents. So we saw another 88 X drop, not 88% or 80, you know, 88, 88 times cheaper in 11 months for that next generation of mod…”
Guo: AI inference clouds are growing 1,000x in compute consumption
“I think to the point of like real data, the inference clouds are growing a thousand X in terms of consumption, right? And then they're getting more efficient. So revenue grows at some lower rate than that, but it's wild.”
Gil: San Francisco housing is rebounding due to AI lab tender liquidity
“SF housing is starting to rise again in part due to I'm assuming outcomes from the lab tenders and things like that. Cause suddenly you have these companies that are worth hundreds of billions of dollars out of nowhere in a few years. And as employees are sell…”
Gil: Top eight tech firms total $23T and over 50% of S&P
“Now the top eight tech companies are about 23 trillion of market cap, and they make up well over 50% of the S&P in terms of value. At the same time they went from basically four percent of GDP in 2005 to about 12% of GDP today.”
Gil: Tech could reach 15% to 30% of US GDP by 2035
“And AI is a driver of this, right? Because you're taking services and you're taking certain types of jobs and you're augmenting them with AI and you're converting them into effectively software spend or tech spend. And you can make different assumptions about …”
Gil: Most Startups Have a 12-Month Peak Valuation Window Before Crashing
“Because often for companies, there's about a 12 month window where your company's the most valuable will ever be, and then it crashes out. For a very small handful of companies, the answer is you should never, ever, ever sell. For most companies, the answer is…”
Gil: Founders Should Pre-Schedule Bi-Annual Board Meetings on Exits
“And one tip for founders is from a hygiene perspective, but also just a way to make it a non-emotional discussion is pre-schedule once or twice a year, the board meeting where you talk about exits.”
Gil: Building a bundle is software's best defense against AI shifts
“The best way to defend against this is to build a bundle.”
Gil: AI compresses decade-long tech displacement cycles into two years
“And so the difference with AI, Is the velocity of change is so high that what normally would have taken a decade, and you'd have a normal decade-long displacement cycle, is now happening in a year or two. And that's really the reason that these things are so t…”