Industry executives and experts share their predictions for 2026. Read them in this 18th annual VMblog.com series exclusive.
By Tommy Levi, Senior Director of AI/ML at Agiloft
We’re living through one of those rare moments when an entire industry recognizes that everything is about to change due to artificial intelligence. But here’s my contrarian take as someone who leads AI development in the contract management lifecycle (CLM) space: while everyone’s racing to implement AI, they’re missing the bigger opportunity sitting right in front of them.
In 2026, the organizations that win won’t be the ones with the flashiest AI demos. They’ll be the ones who figured out how to unlock the intelligence trapped in their workflows.
The AI Paradox: Powerful Tools, Untapped Data
I’ve spent my career cutting through AI hype to focus on what actually solves customer problems. And right now, I’m watching a fascinating paradox unfold: companies are making massive AI investments while simultaneously ignoring one of their richest data sources.
Your contracts contain the DNA of your business relationships-pricing terms, obligations, renewal dates, risk provisions, termination clauses. Every negotiated point represents a business decision, every obligation is a commitment that needs tracking, and every risk provision is a potential exposure your finance and legal teams should monitor in real-time.
Yet for most enterprises, this data remains locked away in document repositories, accessible only through manual review or basic keyword search. It’s the equivalent of having a gold mine in your basement and never bothering to dig.
Why 2026 Will Be Different
Several converging trends are about to change this equation fundamentally.
- First, AI models have reached a maturity point where they can reliably extract and structure contract data-but only when built on the right foundation. The key word here is “reliably.”
Anyone can build a demo that looks impressive in a 15-minute presentation, but when you’re dealing with mission-critical business data, accuracy isn’t optional. The possibility of delivering misinformation at scale should terrify us, whether we’re talking about business operations or geopolitical decisions.
The enterprises that succeed in 2026 will be those using AI models trained on trusted, structured contract data rather than general-purpose tools that sound confident (while being confidently wrong).
- Second, cloud infrastructure has evolved to a point where contract data can finally integrate seamlessly with existing business systems while maintaining security and compliance requirements. Not only is this about cloud storage, but it’s also about creating an ecosystem where contract intelligence flows directly into the tools your finance, procurement, and operations teams already use.
- Lastly (and this is crucial), customer problems are finally driving the conversation instead of technology capabilities–as they should. For too long, the AI industry has operated on a “if you build it, they will come” mentality. But that’s not how lasting business value gets created. You can’t just slap AI onto a product and expect customers to find uses for it. Engineering teams must start with the workflows that are broken, the decisions that lack data, the risks that go unmonitored-and figure out how AI can solve for those specific problems.
From Silos to Intelligence Networks
Here’s what the shift will look like in practice: imagine your procurement team preparing to negotiate supplier renewals. Instead of manually reviewing past contracts and emailing colleagues for context, they query a system that instantly surfaces all relevant pricing history, identifies favorable terms from similar agreements, flags obligations that weren’t met, and compares current market conditions against existing terms.
Or consider your finance team closing a quarter. Rather than chasing down contract values and revenue recognition schedules, they have real-time visibility into all financial obligations, automatically extracted and kept current as amendments occur.
This isn’t science fiction-the technology exists today. What’s been missing is the infrastructure and intentionality to treat contract data as priority number one in the enterprise data ecosystem.
The Hard Truth About AI Implementation
Let me be direct about something: not every problem needs an AI solution (shocking coming from an AI expert, I know). The fundamentals of CLM-uploading, organizing, tracking, analyzing-aren’t going away, but these basics need to work flawlessly before we layer on sophisticated AI capabilities.
Some of our customers are using AI features extensively and seeing tremendous value. Others aren’t ready yet, which doesn’t necessarily mean that the AI isn’t working. It could be because their underlying data and processes aren’t mature enough to support it. That’s okay. It’s actually healthy! The worst thing we can do is promise that AI will magically solve every problem without doing the foundational work first.
(We’re still waiting for those self-driving cars, after all.)
What This Means for 2026
Next year, we’ll see a clear divergence between organizations that view AI as a feature to check off and those that understand it as part of a broader data and product strategy. The winners will be companies that:
- Invest in extracting and structuring their contract data with models that prioritize accuracy over impressiveness
- Integrate contract intelligence into existing business workflows rather than creating new systems people must remember to check
- Focus relentlessly on solving specific customer problems vs. building AI for AI’s sake
- Maintain realistic expectations-understanding that AI is a powerful tool, not a perfect oracle
The opportunity ahead of us isn’t just about smarter contract management. It’s about transforming contracts from static documents into dynamic intelligence that powers real-time business decisions. That’s the frontier worth pursuing. And unlike most AI predictions, this one doesn’t require any technological breakthroughs. It just requires organizations to recognize the value of what they already have.
##
ABOUT THE AUTHOR
Dr. Thomas Levi is the Senior Director of AI and ML at Agiloft, where his teams focus on building production features for customers leveraging advanced techniques in machine learning, natural language processing and generative AI. He began his professional career as a string theorist in physics, working at NYU and UBC. He then moved into industry, working as a Senior Data Scientist at Plenty of Fish. Since then, he has focused on building research and production Data Science and Machine Learning teams in various SaaS companies, including Unbounce and Fullstory.
