Public discourse on AI often misses a key dichotomy. While consumer-facing AI products are widely disliked and fail to deliver value, AI has found significant product-market fit within the enterprise for tasks like coding and business process automation. This explains the disconnect between venture capital hype and public skepticism.

Mainstream consumers are not actively seeking out AI products the way they did smartphones. Instead, mediocre AI features are being "foisted upon them" within existing apps like Google Search, leading to a perception of low quality and annoyance.

Data from RAMP indicates enterprise AI adoption has stalled at 45%, with 55% of businesses not paying for AI. This suggests that simply making models smarter isn't driving growth. The next adoption wave requires AI to become more practically useful and demonstrate clear business value, rather than just offering incremental intelligence gains.

OpenAI's ambition to create an AI device faces the same challenge that defeated Windows Phone: ecosystem lock-in. Without seamless integration of essential apps like Gmail and Instagram, a new device cannot realistically replace the smartphone for consumers, regardless of its AI capabilities.

Enterprises have immense excitement and budget for AI but struggle to define concrete applications. When asked for discrete use cases, the responses are wildly varied, revealing a "blank canvas" problem. The solution is to meet users where they are with specific, guided applications rather than an open-ended tool.

The slow adoption of AI isn't due to a natural 'diffusion lag' but is evidence that models still lack core competencies for broad economic value. If AI were as capable as skilled humans, it would integrate into businesses almost instantly.

Despite the hype, AI's impact on daily life remains minimal because most consumer apps haven't changed. The true societal shift will occur when new, AI-native applications are built from the ground up, much like the iPhone enabled a new class of apps, rather than just bolting AI features onto old frameworks.

Ramp's AI index shows paid AI adoption among businesses has stalled. This indicates the initial wave of adoption driven by model capability leaps has passed. Future growth will depend less on raw model improvements and more on clear, high-ROI use cases for the mainstream market.

Despite massive spending and partnerships, Microsoft, Amazon, Apple, and Meta have failed to launch a defining, consumer-facing AI product. This surprising lack of execution challenges the assumption that incumbents would easily dominate the AI space, leaving the door open for native AI startups.