The AI Revolution Is Already Here — It's Just Hiding in Broken Workflows

I stumbled on this piece from Lenny Rachitsky about Applied Intuition, a $15 billion AI company working on cars, tractors, and other vehicles. Qasar Younis, the CEO, makes a compelling case that the biggest AI revolution will happen in mining, farming, construction, and trucking over the next 5 to 10 years—not in software. It's a bold vision, and he's probably right about the long-term potential. But here's what the article misses: the AI revolution in these industries is already happening, and it's not about autonomous vehicles. It's about fixing the broken, everyday workflows that cost businesses millions.

Our data at PainSignal tells a different story—one that's less about futuristic tractors and more about real, solvable problems. We're tracking 89 problems in construction alone, with an average severity of 4.3 out of 5. Six of those are severity 5/5 issues related to payment protection and quality control. In manufacturing, we've got 20 problems with an average severity of 4.2/5, including sales-engineering misalignment and FDA documentation disorganization. These aren't sci-fi challenges; they're operational inefficiencies that AI can tackle today. While Qasar talks about a 5-10 year horizon, our data suggests the low-hanging fruit is ripe for picking now.

Take construction, for example. The article mentions AI in physical industries, but it glosses over the specific pain points. We see construction workers struggling with payment protection—six problems with severity 4-5/5. That's not a futuristic AI problem; it's a cash-flow issue that an app could solve tomorrow. Or quality control: two problems with severity 5/5. An AI-powered inspection tool could catch defects before they become costly rework. These are concrete opportunities, not speculative bets. If you're a builder or an investor, this is where the ROI is immediate. You don't need to wait for autonomous bulldozers; you can build a payment platform that uses AI to verify work and release funds, or a quality control app that analyzes site photos for compliance issues.

Qasar argues that the AI revolution in these industries will happen 'not in software,' but our data challenges that framing. Of the 45 app ideas we track in construction, most are software-based solutions that bridge physical operations with intelligent systems. Think about it: a construction foreman isn't going to code an AI model for a tractor, but they might use an app that automates invoice processing or tracks equipment maintenance. The revolution is software-enabled. It's apps that make physical work smarter, not just hardware that replaces it. We're seeing this pattern across industries—from manufacturing workflow tools to logistics scheduling platforms. The real opportunity isn't AI vs. software; it's AI in software that solves real problems.

Another point from the interview that deserves a reality check: Qasar says the most successful companies show traction very early, based on his Y Combinator experience. That might hold true for SaaS startups, but in physical industries like construction and manufacturing, traction can be slower due to industry inertia. Our data shows many successful app ideas address deeply entrenched problems—like payment protection or compliance—where adoption takes time because you're changing decades-old habits. Early traction is great, but in these verticals, solving a high-severity problem (even if it takes a year to gain users) can lead to massive success. Don't let the 'move fast' mentality blind you to industries where trust is built slowly.

So, what does this mean for you? If you're a vibe_coder looking for a project, skip the autonomous vehicle hype and build something that fixes a real workflow. Our construction industry page shows 89 problems waiting for solutions. Or check out the manufacturing opportunities—20 problems with an average severity of 4.2/5. For indie_hackers, this is a goldmine: these are problems people will pay to solve today, not in a decade. And seed_investors, look beyond the flashy AI demos; the data reveals patterns of inefficiency that signal ripe markets. We're tracking 2292 total problems across 92 industries, and many of them are in physical sectors where AI can drive immediate value.

Lenny's interview is worth a read for the big-picture vision, but don't let it distract you from the ground truth. The AI revolution in physical industries isn't a distant future—it's already unfolding in the gaps between broken processes. While Applied Intuition builds intelligence for vehicles, there's a whole ecosystem of apps needed to make those industries work better. And that's where the real building happens.