“The most overlooked opportunity in AI isn’t building an AI startup; it’s using AI to run the unglamorous, real-world businesses everyone else ignores.”
We live in a world that increasingly wants everything to be black or white. You’re on one side or the other, you agree or you disagree, you’re with us or against us. But real life is rarely that simple.
In reality, there’s plenty of room for nuance, contradiction and opinions that don’t fit neatly into one camp; whether we like it or not. Yet, somewhere along the way, we’ve become rather uncomfortable with the messy middle.
That’s the thinking behind Quite Contrary, TechRound’s series dedicated to hot takes, unpopular opinions and ideas that might make you shift slightly uncomfortably in your seat. Each edition starts with a deliberately contrarian viewpoint, then opens the floor for different perspectives.
Because perhaps we don’t always need to pick a side. Sometimes, we just need to have the conversation.
More from Artificial Intelligence
- FlowX.AI Brings Mission-Critical Industry AI Agents To Gemini Enterprise
- AI May Not Replace Your Doctor, But It Will Change What You Need Them For
- What Is an AI Agent And Why Should Your Business Be Using One?
- Meet The AI Agent Doing The Job Of An IT Technician
- OpenAI And Replit Join Forces As The AI Coding Race Heats Up
- Slack Launches AI Coding Channels – Is Your Entire Company A Dev Team Now?
- AI Can Process Medical Data. But Does It Actually Understand How Our Bodies Work?
- The Death Of The Click: Google Gave Us The Click, Is AI Taking It Away?
This Week’s Contrarian
This week, our “unpopular opinion” comes from Samer Bejjani, Co-founder of Shootday. Bejjani believes that the most overlooked opportunity in AI isn’t building an AI startup. Rather, the most overlooked opportunity is using AI to run the unglamorous, real-world businesses everyone else ignores. To do the less flashy things that make the world keep turning.
Samer Bejjani, Co-Founder of Shootday
Samer Bejjani is co-founder of Shootday, a global photography and videography company. He has also co-founded several other ventures, including It Was AWESOME, Growth Levers and Food Label Maker, and holds a Master’s in Computer Software Engineering.
Samer’s Hot Take
“The most overlooked opportunity in AI isn’t building an AI startup, it’s using AI to run the unglamorous, real-world businesses everyone else ignores.”
“Everyone is racing to build the next AI product. I think the most overlooked move is to point AI at the boring businesses that already have customers, revenue and real-world problems.
“I am a co-founder at a global photography and videography platform: a physical, human service, with crews, cameras and shoots across dozens of cities. My co-founder runs the company; I build and scale the technology behind it. Our entire tech team is two people. Our competitors have dozens of engineers. One of the reasons we can compete is that I ship most of our product myself using AI, without a classic engineering background, by treating a messy real-world service as a software and operations problem.
“The lesson I keep learning is that AI’s real leverage shows up when it is aimed at offline, unglamorous operations like logistics, scheduling, pricing and delivery, not at another chatbot competing with a thousand others. The wrappers will get commoditised. The companies that quietly wire AI into real-world businesses will still be here in five years, and most of them will never call themselves AI companies.”
Here’s What the TechRound Community Had To Say
- Kadan Stadelmann: Co-Founder at Compance
- Heath Squier: Founder at CAIO at EVKII
- Ritty Quin: CEO of KOR
- Priyank Jain: Data Scientist II at Boost Mobile
- Andy Gibbs: Founder and CEO, Trusted Remodel Advisor LLC
- Sudhakavya Bodapati Venkata: DevOps Engineer and Researcher
- Achiya Cohen: Founder at Achiya Automation
- Dino Correia: Founder at Agaya Cloud
- Vladimir Beskorovainyi: Enterprise AI Architect and CTO, Founder of Besk Tech
- Chase W. Hughes: Three-Time Founder
- Yang Fang: CEO and Founder of BeagleTechnology
Kadan Stadelmann, Co-Founder at Compance
“Organizations that figure out how to incorporate AI into real-world operations will see measurable gains across disciplines so long as they take an all-encompassing approach to implementing the technology to augment human capabilities rather than replace humans. AI is already proving beneficial in logistics, scheduling, pricing, demand forecasting, warehouse efficiency, and cost reduction.
“Adoption lags due to integration barriers, skills gaps, and data readiness. Pure AI wrappers face rapid commoditization, while firms that embed AI into physical, customer-facing businesses create more durable advantages.”
If AI is as transformative as we all claim, why are so few founders using it to quietly fix boring, real-world businesses instead of building yet another AI tool?
“There are many hurdles faced when it comes to incorporating AI. Many workers are not trained in AI, for instance. Meanwhile, organizations have disorganized data or a lack thereof altogether. Organizations can be resistant to change, especially one with such fear around it like AI does. In particular, around people being concerned for their own jobs. Venture Capitalists are also easily distracted and might be blind to the more mundane tasks that can be automated by AI.”
Heath Squier, Founder at CAIO at EVKII
“Somewhere between agree and strongly agree. AI creates more durable value when it removes friction from an existing operation than when it merely adds another interface. In my earlier food-industry work, product development required taste panels, stability testing, lab verification, packaging decisions, and coordination across a team of more than 70 people. Those workflows were valuable precisely because they were messy, physical, and accountable to real customers.
“Today I use agents to build websites, generate and qualify leads, optimize advertising cost per acquisition, and delegate repeatable work, while retaining human accountability for claims and decisions. The defensible advantage is not access to a model; competitors can buy that. It is encoding a company’s operating knowledge, connecting automation to source, leads, conversions, revenue, and ROI, then learning faster. The “boring” businesses are attractive because their problems are concrete, the baseline is measurable, and efficiency gains compound inside an established customer relationship.”
If AI is as transformative as we all claim, why are so few founders using it to quietly fix boring, real-world businesses instead of building yet another AI tool?
“Operations are less glamorous, require domain knowledge, and force accountability for messy exceptions. Building a tool is easier to demo; changing a real workflow requires trust, integration, and patience.”
Ritty Quin, CEO of KOR
“I agree, but would not draw a clean line between ‘AI startups’ and ‘real-world businesses.’ The strongest AI companies will be those that disappear into a genuine workflow and solve a persistent problem. In entertainment, creation gets attention, but much of the industry still runs on fragmented rights records, manual licensing, slow payments and decisions across disconnected systems. Those are not glamorous problems, but they determine whether creative work can become a sustainable business.
“AI wrappers with no proprietary context, distribution or workflow advantage will be difficult to defend. But applying AI to an existing industry is not automatically valuable either. The company still needs domain knowledge, trusted relationships and a clear reason for customers to change. The opportunity is to understand that business deeply enough to redesign the parts people have accepted as inefficient for far too long.”
If AI is as transformative as we all claim, why are so few founders using it to quietly fix boring, real-world businesses instead of building yet another AI tool?
“Few founders are doing this because building another AI tool is easier to explain, fund and market. Fixing an established industry requires patience, domain expertise and a willingness to confront messy workflows and human incentives. The technology moves quickly, changing how real businesses operate does not.”
Priyank Jain, Data Scientist II at Boost Mobile
“Agree, and I’d go further: the AI gold rush is mostly people building shovels nobody needs. I point AI at the unglamorous stuff, forecasting, location decisions, churn, the boring plumbing of retail and telecom, and that’s exactly where the leverage hides. A model that quietly saves a logistics team real money will outlast a thousand identical chatbots. The wrappers get commoditised in a year; what doesn’t is AI tangled into your actual operations and data, because that’s the part competitors can’t copy.
“One caveat, though: boring businesses bite back. A chatbot that hallucinates is embarrassing. A pricing or scheduling model that’s wrong costs real money. So the winners won’t just bolt AI on, they’ll pair it with serious validation and human judgment. Point AI at the boring stuff, absolutely, but respect how little room for error those businesses give you. The quiet players win the decade.”
If AI is as transformative as we all claim, why are so few founders using it to quietly fix boring, real-world businesses instead of building yet another AI tool?
“Because a demo gets applause and funding; a better scheduling system gets a shrug. The whole incentive stack, status, VC money, exits, rewards the shiny tool, not the slow, domain-heavy grind of fixing something boring. So founders chase the thing that looks like the future. The ones who ignore the applause and fix the boring stuff will still be here when the wrappers are gone.”
Andy Gibbs, Founder and CEO, Trusted Remodel Advisor LLC
“I agree with the destination, but AI alone isn’t the map. Its greatest leverage is in unglamorous businesses, and few are more fragmented than remodeling. It still runs on gut instinct, incomplete estimates and homeowners who don’t know which questions to ask. I treat that mess as a software and operations problem.
“I spent 50 years in construction and hold over 40 patents (including patents in early 2000s Semantic Analysis engines). Now my agentic AI adviser delivers my expertise, answers homeowners’ questions and turns conversations into customers. It does work that once required multiple people / skillsets.
“A chatbot alone is the commodity your argument warns about. The moat is the proprietary knowledge, operating data and commercial system wired around it – not the model. AI can’t manufacture 50 years of domain judgment.
“The wrappers will disappear. Domain expertise embedded in a real business survives. Most winners won’t call themselves AI companies. I don’t. I’m a builder who finally has leverage.”
If AI is as transformative as we all claim, why are so few founders using it to quietly fix boring, real-world businesses instead of building yet another AI tool?
“Because fixing a boring business requires knowing that business. It’s easier to build an AI tool from a laptop than to acquire decades of operational judgment. “AI startup” also attracts attention and capital, but quietly improving remodeling operations does not. The winners will be industry domain experts who build with AI, not AI natives searching for an industry.”
Sudhakavya Bodapati Venkata, DevOps Engineer and Researcher
“AI creates the most durable value when it improves real operations—but only when it is constrained by evidence and accountability. In medical-device manufacturing environments, the important applications are rarely glamorous: detecting failures, correlating telemetry, protecting deployments, reducing recovery time and preserving audit trails. These systems affect product quality and business continuity, so “move fast and break things” is the wrong model.
“My research evaluated evidence-gated autonomous incident response across 175 incident episodes. Requiring predefined evidence before state-changing actions reduced unnecessary remediation and action-explosion events while maintaining high recovery success. That result captures the real opportunity: AI should not replace operational discipline; it should make disciplined teams faster and more reliable.
“The winners will be companies that quietly embed AI into established workflows with least-privilege access, rollback controls and human approval for consequential actions—whether or not they ever describe themselves as AI companies.”
If AI is as transformative as we all claim, why are so few founders using it to quietly fix boring, real-world businesses instead of building yet another AI tool?
“Real-world operations are messy, difficult to integrate and carry genuine consequences. Another AI tool is easier to demonstrate and fund; improving an established operation requires domain knowledge, trust, accountability and patience.”
Dino Correia, Founder at Agaya Cloud
“Technologies are changing very fast, and it is very difficult for SME businesses to keep up with the fast changes in these same technologies. Your business doesn’t need to be necessary to be on the latest technology or developing AI tech all the time.
“What you must have in your business structure is the infrastructure ready to pivot and the ease of implementation of these technologies. If you leave this for a long time, you will be left behind. Sometimes the best way to optimise solutions is to outsource those tools to recognised companies.
“The type of users we’re seeing today are more demanding of the services we provide, as they can easily switch to other competitors due to the release of many platforms that are being developed. The key is to provide excellent customer service and make sure the customer feels valued.”
If AI is as transformative as we all claim, why are so few founders using it to quietly fix boring, real-world businesses instead of building yet another AI tool?
“The new AI coding agents have allowed many people the ability to develop new tools without any prior coding experience. However, scaling a product in production becomes a real challenge for “vibe coders”. That’s when the AI tool developed by companies that are not tech-focused doesn’t scale. Focusing in integration into their business process is more scalable than developing your next chatbot.”
Vladimir Beskorovainyi, Enterprise AI Architect and CTO, Founder of Besk Tech
“The claim is that founders overlook these businesses. They do not. They avoid them, and rationally. Wiring AI into logistics or scheduling means learning logistics or scheduling, and that knowledge does not transfer to the next customer. A wrapper takes a weekend and sells to everyone. One is a job, the other is a product, and founders are responding to that difference rather than missing it.
“Which is exactly why the opportunity is real. The barrier is not technical, and no better model clears it.
“The part I would push back on is durability. What protects a two-person team is not the AI. It is the written description of how the business actually works, and that starts decaying the day it is written. Whoever keeps it true stays in business.”
If AI is as transformative as we all claim, why are so few founders using it to quietly fix boring, real-world businesses instead of building yet another AI tool?
“Because it is not a startup, it is a job. The work is learning somebody else’s business, it does not compound across customers, and it cannot be pitched as a category. Founders are selecting for what is fundable rather than for what is valuable.”
Chase W. Hughes, Three-Time Founder
“Agree with the conclusion and not with the reasoning.
“I did this. My first company was a consulting firm: manual, human, deeply unglamorous work. In 2021 I fine-tuned its data on GPT and built ProAI on top of it – 300,000+ users, bootstrapped, about eighteen months to an exit. So yes, the boring business was the advantage.
“But not because boring businesses are soft targets for AI. Because they hand you two things a wrapper never gets: proprietary data nobody else can buy, and a ground-truth feedback loop. When the work is wrong, a customer tells you that week. That correction signal, not the model, is the moat.
“Where I part company: offline operations are the hardest place to run AI, not the easiest. A bad chatbot answer is visible immediately. A scheduling agent that reports success and quietly did nothing is not.”
If AI is as transformative as we all claim, why are so few founders using it to quietly fix boring, real-world businesses instead of building yet another AI tool?
“Because that work cannot be demoed. An AI tool ships in a weekend and gets a launch post; fixing dispatch for a company with crews in thirty cities takes six months of learning how dispatch actually breaks, and the reward is an invoice rather than a headline. Founders follow the shorter feedback loop. The ones already doing it are also invisible by design – they do not call themselves AI companies, so the pattern never propagates and the next founder never sees it.”
Yang Fang, CEO and Founder of BeagleTechnology
“I agree. The biggest opportunity for AI is not necessarily building yet another “AI-native startup”, but instead ushering that intelligence into the physical world.
“There are millions of businesses still dependent on repetitive manual processes, legacy equipment, and workflows that have changed very little (if at all) over the past few decades. Manufacturing, logistics, construction, agriculture, warehousing, and field services don’t always look exciting from the outside, but they represent enormous parts of the global economy.
“Physical AI creates an opportunity to change how these businesses operate from the inside out. When AI is combined with machines, sensors, vision systems, and automation, it gets the opportunity to stretch its legs and move past generating information and actually perform useful work in the real world.
“The companies that take AI out “for a spin”, deploy it reliably in difficult environments, and solve problems customers are willing to pay for —will be the real winners in the AI boom.”
If AI is as transformative as we all claim, why are so few founders using it to quietly fix boring, real-world businesses instead of building yet another AI tool?
“I think founders naturally chase where the attention and capital are headed. Boring, real-world problems aren’t that “exciting” at first, like high-growth opportunities, and “real-world” problems are typically much harder to solve. But when push comes to shove, so many of the AI products we see on the market are ultimately nice-to-have solutions, not must-have solutions. The overlooked opportunity here is using AI to solve problems that businesses must solve, even if those problems aren’t deemed exciting or making the most headlines.”
