The Bacon & the Skillet: When Does the AI Market Congeal?
Автор сравнивает текущий рынок ИИ с шипящим беконом на сковороде: всё бурлит и стремительно меняется. Доли рынка подвижны, потому что модели постоянно улучшаются — производительность чипов Nvidia и запуск Gemini 3 (крупнейший скачок в истории моделей Google) не предвещают замедления. Пока модели приближаются к уровню PhD, пользователи продолжают тестировать и переключаться, ведь выигрыш оправдывает энергию активации. Стартапы, инкумбенты, облачные провайдеры и AI-лаборатории конкурируют на всех слоях: модели, инфраструктура, инструменты, приложения. Но эра текучести не вечна: когда разрыв между лучшей и второй моделью сократится, издержки переключения начнут перевешивать прирост производительности — и жир застынет. Победят те, кто за фазу «шипения» успеет нарастить достаточно жира, вокруг которого стоит застывать.
The AI market today is bacon in a hot skillet. Everything is sizzling, moving, & changing at an incredible pace. We’re all watching it closely.
Market share is fluid because no one yet knows what AI can do & the second we think have grasped it, models improve. The Nvidia chip performance & the launch of Gemini 3 the biggest gain ever in Google model performance suggest no simmering ahead.
As long as the underlying models hurtle towards PhD level performance, people will continue to test. How much better is Gemini 3 at coding? tool calling? writing?
If the progress is material, then the benefit of switching is worth the activation energy.
Today, startups, incumbent software companies, cloud providers & AI labs all are competing. First the model, then infrastructure (memory & retrieval), then tools, then applications. Will the foundational models play at the application layer? Or will the applications differentiate themselves enough to overcome model differences?
Who can take advantage of the next big leap in model performance fastest? Which sales team can reach the target customers first & write the RFP?
This is the Great Game of Risk in Category Creation & aggression wins.
But this era of fluidity won’t last forever. The rate of improvement in AI models will eventually attenuate. When the performance gap between the best model & the second-best model shrinks, the incentive to switch evaporates.
Switching costs will start to matter more than marginal performance gains. The custom tools I’ve built, the muscle memory I’ve developed, the integrations my company has deployed, the enterprise contracts signed, all inertia.
At that point, the fat begins to congeal.
The winners will be those who use the sizzling phase to build fat worth congealing around.
This fun analogy came up during my conversation with Harry, Jason, & Rory.