# The Build-Cost Collapse Moves the Bottleneck to Idea Quality ## Claim When AI makes implementation nearly free, the scarce resource in personal software becomes finding a problem worth solving, not the skill required to solve it. <!-- atomicity: 1 idea ✓ --> ## Scope & conditions - Holds for small, personal, well-scoped utility software — the class of app a non-engineer can vibe-code end to end in a sitting. - Breaks where correctness, scale, or security dominate the cost — there, implementation skill (and its absence) is still the binding constraint, no matter how cheap generation gets. - The panel's own evidence for the scope limit: most output is throwaway ("this is total garbage" on second use), which means idea quality, not build cost, was already the failure point even before security is considered. ## Evidence Three non-engineer builders on The Vergecast independently converge on the same read: Jake Kastrenakes names it directly ("finding an idea that is useful to you specifically is harder than building it now"), and Hayden Field's parallel observation — vibe-coded apps work best on problems that are "really easy, but you just can't get it together" to solve manually — describes the same shift from a skill bottleneck to a motivation/discovery bottleneck. ## Links - [[AI & the Future of Knowledge Work MOC]] — the relocation of scarcity from implementation skill to knowing what's worth building - [[3. Resources/Permanent Notes/Personal Fit and Shareability Trade Off in Vibe-Coded Software]] — because extends; once idea-finding is solved, the next unsolved problem the same source names is making the resulting app safe to share. - [[3. Resources/Permanent Notes/AI's Leverage Comes From the Data Foundation Not the Model]] — because contrasts; that note locates AI's ceiling in data quality for organizational decision-making, this one locates it in human idea-selection for personal tools — different domains, same "the model isn't the bottleneck anymore" shape. - [[3. Resources/MOCs/Product & Startup Instincts MOC]] — because applies-in; if implementation is cheap for everyone, product instinct (knowing what's worth building) becomes the actual differentiator. ## Tests / falsifiers - Track a cohort of new vibe coders over a month: if the apps they abandon fail primarily on execution bugs rather than "I don't need this after all," the claim is wrong for that cohort. - Compare idea-generation time vs. build time across projects — if build time still dominates for typical personal-utility scope, the "build cost collapsed" premise itself needs revisiting. ## Maintenance - Last tended: 2026-08-16 - Next review: 2026-11-16 ## Source - [[3. Resources/Readwise/Articles/Our Vibe Coded Projects That Actually Work the Vergecast]] — Jake Kastrenakes and Hayden Field's on-air observations about the shifted bottleneck