What Trellis Impact 26 Signals for AI x Sustainability
Key Summary
- AI was everywhere at Trellis Impact 26, shaping sustainability discussions beyond the dedicated track.
- Real use cases are live, from risk modeling and supplier data capture to reporting and siting analysis.
- Human verification still matters, as AI outputs can be inauthentic or not fit for purpose.
by Kelly Scanlon, senior director of sustainability strategy, Global Electronics Association
I attended Trellis Impact 26 on behalf of the Global Electronics Association to deepen our understanding of how AI is shaping sustainability practices, the infrastructure behind it, the software powering it, and how various industry sectors, including the electronics sector, are wrestling with both the promise and the problems AI presents. What I didn't expect was how thoroughly AI had saturated every conversation at the event not just the sessions dedicated to the topic.
It was the question at the workshop tables, the follow-up during Q&A, the thread running through exhibitor demos and hallway side conversations. What agents and models are you using? What are you prompting them with? How do you check for accuracy? AI isn't just a topic anymore. It's the lens through which we are all starting to see everything.
The AI x Sustainability Track Highlights Real Use Cases
The dedicated AI x Sustainability track gave startups a platform to show what's actually being built and deployed using AI. I saw presentations about how AI is being deployed to assess fire risk and electrical grid vulnerabilities, collect supplier data through simple email exchanges and organize it for sustainability reporting, evaluate land use for data center siting, and analyze temperature readings at two meters above ground generating the kind of granular climate data that used to require significant manual effort to process. These aren't hypothetical use cases. These are products drawing real interest from partners and investors in the room and in the electronics industry.
"This Is New to Everyone, Isn't It?"
One of the most honest moments was in one of the first sessions of the conference event when someone prefaced their AI-related question with, "I'm new to this.” They didn't need to say it. Nearly everyone is new to this. And yet there's a tendency to feel like you're behind or like everyone else has figured out the prompt, the workflow, the verification process. You're not. This is genuinely new territory for everyone. Isn’t it?
There was something grounding about watching exhibitors, panelists, and attendees openly compare notes in real time; what's working, what's not, as well as where AI adds real value, and where it still needs a human check.
When AI Isn’t the Right Fit
As an anecdote to the rather intense AI-related questions and conversations happening, one attendee was trying to get a real photo from the conference photographer. He didn't like his AI-generated headshot. He wanted something more authentic.
Amid all the enthusiasm for what AI can generate, there's still a meaningful preference for what's human, -- for the thing a model can only approximate but not quite replicate. That tension isn't going away, and for an industry thinking seriously about AI's role across the value chain, it's worth noting.
When AI Does your Homework Before You Ask
I returned from the event to find an unsolicited email in my inbox. A company managing sustainability and utility data had used their AI system to generate a benchmarked energy report about the Global Electronics Association for free, as a marketing tool. It had pulled building data, cross-referenced public disclosures, layered industry-peer benchmarks, and added market intelligence.
The sample report wasn't accurate for us because as an industry association, our sustainability obligations look different from an electronics manufacturer's. We appreciated the effort and are impressed by the gesture, but it’s a warning that AI can be used to gather insights that might not be the right fit or give you meaningful answers.
Still, it told a story: a complete, benchmarked sustainability report generated automatically, delivered before you've even unpacked your conference bag.
What's clear is that AI is not peripheral for the global electronics industry; this industry is not a bystander to AI. AI is going to be central to how to set and meet sustainability targets like net zero goals, manage supply chain complexity, and manage regulatory pressures. The conversations at Trellis Impact 26 confirmed that questions about AI's environmental, economic, and social implications are being asked right alongside questions about what AI can do.