AI in EHS&S: Seismic Shifts and Untapped Opportunities
Luke Jacobs
March 31, 2026
What’s Holding Companies Back
Despite this seismic shift in what AI can accomplish, the data problem remains daunting for many companies. There’s still a high volume of compliance-critical information being generated in handwritten logs, disconnected spreadsheets, and scanned PDFs. And AI can’t fix a garbage-in, garbage-out problem.
Layered on top of that is a change management challenge: the people who know the most about a facility's compliance posture are often the most skeptical of technology and AI tools. But getting their buy-in isn't optional. These people are also often extremely busy, so assuming they have the internal bandwidth to become AI power users and develop internal tools is unrealistic.
Untapped Opportunities Abound
The good news is, where there are challenges there are also opportunities. One worth talking about is cost replacement potential in data-heavy manual work. Not as a cost-cutting measure, but as a fundamental restructuring of how environmental compliance work gets done. A staggering percentage of EHS spend flows to environmental consultants and internal teams performing high-volume, rules-based work: DMRs, Tier II reports, RCRA filings, permit applications, annual emissions inventories, etc. These are exactly the workflows that well-architected AI agents will be able to execute at a fraction of the cost and in a fraction of the time. The companies that recognize this early have an opportunity. They can redeploy that spend and internal time toward strategic work rather than just pocketing the savings.
Beyond better spend utilization, the deeper opportunity is data infrastructure: companies that invest in clean, centralized chemical, waste, air, water, and generalized facility data systems right now will have a compounding advantage. Every new regulation that comes down the pipeline will be something they can respond to in days rather than months, because they're not starting from scratch.
And perhaps most underexplored is the connection between compliance data and sustainability strategy. The chemical inventories, emissions records, and waste manifests that exist purely to satisfy regulators today are the exact same data needed to set credible reduction targets, track progress, and tell a coherent sustainability story to investors. The companies that see those as the same dataset, not two separate functions, are going to be structurally ahead.
What’s Working (and What Isn’t)
The teams using AI well share a common discipline: they start narrow. Pick one high-frequency compliance workflow, prove ROI in a defined timeframe, and expand from there. Implementation itself can be used as a forcing function to finally fix their underlying data. They’re also good at distinguishing what can be improved in-house and what is better off being built and maintained by a specialized vendor. With mission critical systems like those governing compliance, companies are likely better off using vendors specialized in those domains compared to internal IT teams that can deliver a short-term project, but don’t have the internal resources to get into every edge case and then maintain that system in perpetuity.
On the flip side, assuming general-purpose AI systems like Copilot, ChatGPT, or Claude will perform well on EHS tasks because the regulations are public fails quickly. Environmental compliance is hyperlocal, context-specific to each location based on their operations, and full of jurisdictional exceptions that require domain-trained models and human expert validation that are maintained in perpetuity, not a large language model pointed at a PDF of the Clean Air Act.
Lastly, companies can make the mistake of buying AI-branded software that only automates a small amount of the compliance process — formatting a report, generating a summary — while leaving the hard parts entirely manual, either for internal teams or consultants.
How to Move Forward & Make Progress
Make the mindset shift on which everything else depends. Stop treating EHS data as a compliance cost center and start treating it as a strategic asset. The companies positioned to win through the next wave of regulatory change are the ones building data infrastructure today, not because regulators are forcing them to, but because they've decided to get ahead of it.
Join Luke and other AI thought leaders at OPEX/TECH26, April 14-16, in St. Petersburg, Florida. Luke is participating in the keynote panel “Unlocking AI-Driven Digital Transformation in EHS & Sustainability” and the session “Strengthen Environmental Oversight with AI-Driven Compliance.”
Related
Sponsored Content
About the Author
Luke Jacobs
Encamp Inc.
Luke Jacobs is the Co-Founder and CEO of Encamp, a software company helping enterprises modernize and automate environmental compliance.