Unlike the cloud, AI sells itself. Business leaders can experience the value firsthand, shifting the conversation from “if” to “how.”
In a recent talk Y Combinator, Box CEO Aaron Levie broke down why this moment is different:
From Convincing to Implementing: The challenge isn’t convincing leaders AI is the future; it’s making it safe, reliable, and trustworthy.
Agents Unlock Hidden Value: AI agents unlock the value of “unstructured data” (contracts, presentations), turning folders into a queryable corporate brain.
Small Teams, Big Reach: Small companies can now act 10x their size, making this a once-in-a-decade opportunity for founders.
Cloud adoption required CFO math and IT persuasion. Moving servers never changed how anyone wrote an email. AI changes the work people see and touch.
AI is fundamentally different. As Box CEO Aaron Levie put it: “The head of marketing can go and play with ChatGPT and be like, ‘Wow, this seems to write marketing copy maybe better than even my own marketing people.’”
No convincing required.
The Shift: From “Will AI Matter?” to “How Do We Trust It?”
Science fiction primed everyone for this moment. From HAL 9000 to self-driving cars, society has been culturally prepared for artificial intelligence. But the ChatGPT moment made it tangible. Business leaders across functions (marketing, operations, finance) could directly experience AI’s capability.
“You don’t need to sell them anymore that AI is like clearly the future,” Levie explained. “Now it’s actually just about like how can you go implement something that’s going to be safe, reliable, works with your data, you can trust it.”
This represents a fundamental shift in enterprise technology adoption. The question isn’t whether AI will transform business. It’s how quickly organizations can implement it responsibly.
The Hidden Asset: Unstructured Data
Most enterprise data lives in formats that resist automation (contracts, presentations, customer communications). Unlike database records with clear fields and relationships, these documents contain knowledge in free-form text that computers couldn’t previously understand.
“AI agents basically thrive on unstructured data,” Levie noted. “All of a sudden all the data that’s inside those folders becomes immensely valuable to companies because now they can ask all that data questions.”
The transformation scope is massive. Marketing teams can query years of campaign materials for insights. Legal departments can automate contract analysis. Research teams can process vast document libraries in minutes rather than months.
For the first time, the majority of enterprise intellectual property becomes computationally accessible.
The Real Work Problem
Most leaders underestimate how much of their workforce is trapped in low-value tasks. When Levie speaks to companies about AI’s potential, “they instantly their eyes light up because they realize, well, now I can actually free up my time and my employees time to go do much more interesting things.”
The reality inside most organizations: employees spend most time on necessary but non-strategic work. Finding information, manual data entry, routine analysis, administrative tasks that don’t differentiate the business but consume resources.
“The vast majority of time inside of a company is on the stuff that really is not strategic,” Levie observed. “If you could free up a company to work on the stuff that’s strategic and not the basically unstrategic stuff that doesn’t differentiate them, most companies actually have a large set of things they would go do with their time.”
AI doesn’t eliminate jobs. It eliminates the work that prevents people from doing their best work.
Press Gets It Wrong
Headlines highlight layoffs. The deeper story is leverage.
Amazon announcing fewer expected employees due to AI efficiency gains misses the bigger picture. Large incumbents may optimize existing operations, but the real opportunity lies elsewhere.
“Imagine the 50 person company where all of a sudden they can act like a 500 person company,” Levie suggested. “Will that company become a 100 person company more quickly than pre-AI? And my argument would be yes.”
Small companies gain disproportionate leverage. They can serve more markets, develop features faster, and provide better customer service while remaining lean. The economic benefits flow to growth rather than just efficiency.
New Business Models Emerge
Companies like Sierra (customer service agents) and Cognition Labs (software engineering agents) already demonstrate this model. They sell outcomes (resolved tickets, completed features) rather than software access.
Traditional SaaS pricing (selling software licenses per employee) breaks down when AI agents can perform unlimited work. A legal software company can’t sell based on lawyer headcount when their agents do the work of hundreds of lawyers.
The shift moves toward consumption and outcome pricing. “Previously a human would cost $5 or $10 per contract to review based on human time. AI agents can do this for 10 cents. So then you charge that customer $2,” Levie explained.
Successful AI companies will layer substantial software on top of raw AI capabilities. Just as cloud storage companies don’t compete solely on storage costs, AI companies won’t compete only on token costs.
Window Is Open (But Won’t Last)
“This window will end. It’ll be over in two or three years from now,” Levie warned. “You’re in the window right now where maybe it won’t be your first attempt, maybe it won’t be your second attempt, but in this window between a year ago and three years from now, this is when the next hundreds of great companies will get started.”
Markets rarely give both disruption and timing clarity. AI does (briefly).
The AI transition creates new categories of work that software can finally address. Professional services, manual processes, and knowledge work that required human intelligence can now be delivered through agents.
But windows close. Incumbents adapt, markets consolidate, and the next wave of technology disruption begins brewing.
What Leaders Should Do Now
For enterprise leaders: Stop debating whether AI will impact business. Focus on safe, reliable implementation. Start with specific use cases where value can be measured. Build internal capability to evaluate AI tools and manage associated risks.
For entrepreneurs: Look for work that’s never been automated because it required human judgment. Find the intersection of professional services and software delivery. Read Innovator’s Dilemma, Crossing the Chasm, and Blue Ocean Strategy to understand market disruption patterns.
For everyone: This moment resembles the early internet or mobile transitions - foundational technologies that reshape how people work, communicate, and create value. The organizations that move thoughtfully but quickly will define the next decade of business.
The AI moment isn’t coming. It’s here. The question is what people will build with it.
September 16, 2025 Aaron Levie: Why Startups Win In The AI Era https://youtube.com/watch?v=uqc_vt95GJg
This analysis draws from Aaron Levie’s recent YC talk and BoxWorks presentations on enterprise AI strategy. For more insights on AI implementation patterns, follow along as we explore vendor perspectives, executive challenges, and practical deployment strategies from enterprise AI’s front lines.