Founder Fridays No. 183
AI Kills Your Interface -- ChatGPT Ads -- Game Your AI Workflow
Happy Friday.
AI Kills Your Interface
The $25K/seat vertical SaaS you built on workflow UX is now worth zero—because when users type plain English into a chat box, years of carefully designed UI muscle memory becomes instantly obsolete. Your competitive moat wasn’t the data (it was licensed or public); it was the interface layer, and AI just made that layer disappear by replacing it with natural language that runs entirely in the cloud. This week, audit which parts of your product are actually “interface tax” versus genuine workflow value—ask your team: “If a user could describe what they want in one sentence to an AI agent, would they still need this feature?” The companies that survive won’t be the ones defending their UI; they’ll be the ones rebuilding around agent-first architectures before their customers realize they’re paying premium prices for what’s become a thin client with an expensive legacy wrapper. Stratechery (8 minutes)
ChatGPT Ads
Ad agencies are flooding into ChatGPT advertising without measurement tools or brand safety guarantees because they’ve watched consumer discovery migrate to AI platforms and know missing this window means losing the next decade of digital distribution. The underlying shift is that search behavior is moving from Google’s ten blue links to conversational AI—and the first movers who figure out how to show up in those AI-generated answers will own customer acquisition before pricing goes through the roof. This week, allocate 5-10% of your paid budget to test ChatGPT ads (or similar AI platform ads) and track one metric obsessively: conversion rate compared to your Google Search campaigns, because you need to learn what messaging works in conversational contexts before your competitors do. If ChatGPT converts even 70% as well as Google at half the CPC, you’ve just found the arbitrage opportunity that funds your next growth phase—but only if you move before everyone else figures it out. AdAge (3 minutes)
Game Your AI Workflow
Treating your AI system like a board game—where you first identify “what are the pieces and what do they do” before optimizing strategy—unlocks building capability that feels impossible when you’re staring at a blank Claude prompt wondering why your AI project stalled out. The insight is that most founders get stuck not because AI tools are inadequate, but because they skip the “teach” phase: defining the discrete components (your meeples, your victory conditions, your resource tokens) before trying to architect the whole system. This week, open a doc and literally list out every tool, prompt, data source, and handoff point in your AI workflow as if you’re writing game instructions for a nine-year-old—give each piece a simple name and one-sentence job description. When you can see your AI system as modular game pieces instead of an overwhelming tangle of prompts and APIs, you’ll spot exactly which piece is missing or broken, and suddenly those “infinite possibilities” compress into 2-3 concrete next moves that actually ship. Every (4 minutes)
Founder FAQ: What are the pros and cons of using Open Source for a startup?
The abandoned npm package you depend on for authentication hasn’t been updated in 18 months, and when it breaks in production, you’ll discover there’s no commercial support to call and your senior engineer just quit—this is how “free” open source becomes a $50K emergency contractor bill at the worst possible moment. The underlying issue is dependency risk: startups treat open source like a buffet, adding packages without checking if projects are actively maintained, have strong communities, or even align with their compliance requirements, then act shocked when a critical library gets abandoned or fails an enterprise security audit. This week, create a simple spreadsheet tracking your top 10 dependencies with three columns: last commit date, number of active maintainers, and whether commercial support exists—flag anything unmaintained for 12+ months or single-maintainer projects as “migration candidates” before they become production emergencies. The payoff is sleeping through the night instead of scrambling when that obscure parser library your entire API depends on suddenly stops working, because you proactively migrated to a well-supported alternative while you still had the bandwidth to do it right. Westaway (4 minutes)
Startup Funding Guides
I’ve put together a series of guides to equip founders to excel at fundraising. These guides break down the deal term-by-term and give you negotiation tips so that you can speak to investors with confidence.
Convertible Note: Guide / Video
Control Legal Spend
Startups suffer from unpredictable legal bills under the billable hour system. Fees fluctuate month to month without warning. Law firms drag out billable hours, but startups foot the bill. Even basic work can lead to surprisingly high legal bills. This unpredictability cripples financial planning. Budgets rarely match actual spend. With utter uncertainty around legal spend, startups cannot forecast or manage burn rates effectively. The antiquated billable hour system fails them. Our General Counsel flat, monthly fee service gives startups cost certainty. Legal spend becomes predictable with bundled services and no surprise overage bills. By switching from hourly to our flat-fee model, startups finally get confident budgeting, accurate forecasting and predictable legal spend. If you’re sick of getting surprise legal bills and are interested in controlling your legal spend, let’s talk.