Your marketing MVP spent the day moving data into slides. Your legal team reviewed this week’s contracts — line by line. Procurement parsed through this week’s vendor onboarding documents (manually).
Sound familiar? That’s because tedious tasks are still alive and well. Even in the AI-first era.
IDC reports that only 26% of companies primarily source automated methods to analyze unstructured data. A recent McKinsey study shows that employees spend 1.8 hours per day searching for information. That’s nearly a quarter of the workweek lost to digital scavenger hunts.
In a recent episode of the “Intelligent Squared” podcast, What Are The Essentials For Reimagining Work with AI Agents?, Box CEO Aaron Levie shared how AI fundamentally reshapes tasks and goes even further to help businesses reimagine entire operations.
Key takeaways:
- Knowledge workers spend nearly a quarter of the week searching for information
- AI agents transform work by completing entire workflows autonomously
- Early AI adopters report 37% average productivity improvements
- Start with experimentation and push AI beyond what teams believe is possible
Knowledge workers have become data entry clerks, trapped in minutiae that AI agents could be handling. Citing an example of physicians spending hours writing up patient notes, Levie shares: “If you're a doctor, and you’re spending your day meeting with patients, the last thing you want to do is spend hours and hours writing up all the notes. That’s effectively homework. You’re not improving the patient experience anymore.”
Most enterprises aren’t pushing AI enough. They’re happy with too-limited gains from AI models — and they could actually be going way bigger.
Think about your team’s workday and how much time they spend:
- Moving information from one document to another
- Manually extracting data from contracts or invoices
- Writing up notes from meetings
- Searching through files to find specific information
- Creating reports by pulling together existing data
Levie posits that it all signals a universal pain point: endless hours in files, moving information around, collating data, reading, copy-pasting, and modifying graphics. “That’s not that enjoyable. Maybe it is the fifth time I do it, but not the seven-thousandth time,” he quips.
The scale of the problem is staggering. Box’s Succeeding in the AI-first era report shows every enterprise now uses 1,500-3,000 SaaS applications, making it impossible to track where all the content lives. The same report states that 81% of IT leaders name data silos as a big blocker to digital transformation.
This isn’t just about individual productivity. It’s about entire organizations built around manual information processing. When your lawyers spend more time reviewing documents than advising clients, or your salespeople spend more time creating presentations than building relationships, you have a systemic problem.
Agents change everything
Tools like ChatGPT work through single interactions. You ask a question. You get an answer.
AI agents represent a fundamental shift. As Levie explains, an AI agent is the idea of a model that can “loop through itself over and over again to complete an entire task.”
Most organizations are starting to get it: 87% have at least begun piloting AI agents, with 63% already deploying intermediate agents and 41% piloting fully autonomous operations. “Now, we’re no longer bottlenecked by human time and our labor,” he says. “Whatever the workflow is, we can free up that time to go do much more strategic, interesting, and important work.”
Levie details the transformative potential: “What you really need is an agent to say, ‘I’m going to go read each individual document. I’m going to pull the most relevant pieces of information. I’m going to keep track of that information in some kind of stored temporary memory. And then I’m going to generate a full report.”
“About 90% of the data in an enterprise is unstructured,” Levie explains. “Yet it’s only useful when humans can actually look at it. That’s vastly more data than a structured dataset. And yet, we actually get so much less out of it because we’re constrained by human time.”
This challenge creates massive bottlenecks. Every contract review, every research synthesis, every report generation requires human time. But here’s the breakthrough: AI agents can now read, understand, and act on unstructured data at scale.
Here’s the breakthrough: AI agents can now read, understand, and act on unstructured data at scale.
Organizations are already seeing AI transformation deliver results, with early adopters reporting 37% average productivity improvements. Immediate benefits include faster information retrieval and analysis, streamlined approvals that eliminate bottlenecks, and 24/7 operations with consistent quality.
This pattern repeats across departments and industries:
- Legal teams shift from document review to strategic counsel
- Sales reps move from slide creation to relationship building
- Finance departments evolve from data extraction to analysis and insights
- Marketers progress from asset management to creative strategy
Action plan: Turn drudgery to strategy
Companies face a choice: Continue treating knowledge workers as expensive data processors, watching talent burn out on tasks beneath their capabilities. Or embrace AI agents and free humans to do what only humans can: Create, strategize, build relationships, and innovate.
To stay competitive, Levie says companies should “bias toward pushing your company to think about doing more: better serving customers, shipping more product, making the business be able to deliver better for employees and customers.”
Transforming starts with equipping teams with resources that elevate their roles. Levie offers three key takeaways for business leaders:
- Your AI strategy is your data strategy. Before deploying AI agents, companies must address their data foundations. As Levie explains: “Most companies don’t have an AI problem. Theyhave a data problem.”
- Start small, but get going right now. Begin with experimentation: “Try many different things across an organization to figure out what’s working, what’s most effective, and then figure out how to scale the things that are working.”
- Push the limits of these AI models far past what you think is possible. Don’t stop at to-do lists. Challenge your teams and your AI agents to think bigger.
Companies that start now — experimenting, learning, building the right foundations — will define the future of work.
“There’s no magic pixie dust,” Levie believes. “But the benefit is that you get a huge burst in productivity because we’re doing a lot less of the drudgery work in that process."
Sounds a lot better than endless copying and pasting.
There’s no magic pixie dust, But the benefit is that you get a huge burst in productivity because we’re doing a lot less of the drudgery work in that process.
