UW researchers built AI agents that quickly estimate electronic devices’ carbon footprints
Our take
Here's an editorial piece fulfilling your requirements, written from the Brand Essence perspective:
This is a solid development, honestly. At UW, they’re consistently pushing boundaries—it’s just part of the vibe. The news of researchers building AI agents to estimate the carbon footprint of electronics is a really practical application of AI, and it speaks to a growing awareness of our collective impact. We’re seeing increasing scrutiny of supply chains and manufacturing processes, and this kind of tool could be a game-changer for both companies and consumers. It's great to see UW celebrating the ingenuity of its students and faculty, just as they did at the recent commencement [UW celebrates Class of 2026 with 151st Commencement in Husky Stadium and ceremonies in the Tacoma Dome and HecEd]. The speed of the estimations—a minute compared to the time it takes human experts—is particularly impressive. It’s a quick update on a complex issue, and that efficiency is huge. It also builds on the broader trend of leveraging technology to tackle challenges, work that’s similarly being explored in fields like dentistry [UW Dentistry researchers testing oral bacteria transplants to cure bad breath]—showing a commitment to innovation across disciplines.
The significance here goes beyond just assessing individual devices. Think about the implications for larger organizations. Companies are under increasing pressure to report on their environmental impact, and this AI system could streamline that process considerably. It could also be used to inform design decisions, encouraging manufacturers to prioritize more sustainable materials and production methods. We're seeing a real shift towards accountability, and this is a tool that actively supports it. It’s not just about feeling good; it’s about making tangible changes. The current challenge, of course, lies in ensuring the accuracy and transparency of the data used to train these AI models. Garbage in, garbage out, as they say. Getting that foundational data right will be key to building trust and driving real change within the tech industry. It’s a beta phase, and expect iteration to come as the models interact with more diverse device types.
The timing of this announcement feels particularly relevant given the ongoing efforts to improve disaster preparedness across the region. The work being done with ShakeAlert installations [With ShakeAlert installations complete, researchers explore offshore expansion] demonstrates a commitment to anticipating and mitigating risks, and this AI tool aligns with that proactive mindset. It’s about identifying potential problems—in this case, environmental impact—and finding solutions before they become critical. It’s a “future me will thank me” kind of innovation – investing in sustainable practices now to avoid bigger headaches down the road. We're at a point where consumers are actively seeking out more sustainable options and demanding greater transparency from the brands they support. This technology could empower them to make more informed choices.
Looking ahead, it’ll be interesting to see how this AI system evolves and whether it can be adapted to assess the carbon footprint of other products. Could a similar approach be applied to the fashion industry, the food sector, or even the construction industry? The potential applications seem endless. The key will be finding ways to make this technology accessible and user-friendly, so that it can be used by a wide range of stakeholders—from individual consumers to multinational corporations. Ultimately, this is about democratizing access to environmental data and empowering everyone to contribute to a more sustainable future. It’s going to be a lowkey revolution, but it’s one we should all be locked in on.
If you shop on Google Flights, you get a quick comparison for different itineraries: One flight’s carbon emissions may be average, while another’s are 14% higher. But if you go shopping for a new laptop, you likely won’t find quick, comprehensible information on different models’ sustainability bonafides, despite the notable environmental impacts of producing and discarding electronics. In part, that’s because understanding a device’s emissions is difficult and time-consuming, even for experts.
University of Washington researchers developed an artificial intelligence system that automatically estimates the environmental impacts of making different electronic devices. The system uses AI agents — programs that perform tasks autonomously — to comb through publicly available data and conduct life cycle assessments, or LCAs. The system achieves an average error rate of 5%-19%, similar to the accuracy of LCAs conducted by experts.
The team published its findings June 12 in Nature Electronics.
“Recent studies have shown that people are willing to pay more for more sustainable devices,” said senior author Vikram Iyer, a UW assistant professor in the Paul G. Allen School of Computer Science & Engineering. “So there’s growing demand for this information. But a phone, for example, is made of hundreds of chips and other components, and producing each of those causes varying amounts of emissions. Since that data isn’t public or sometimes not even measured, human experts can spend days, even months manually gathering information for LCA. Instead we designed multiple AI agents that work together to automatically find this data and produce comparable estimates in about a minute.”
Related
AI agents have recently grown increasingly capable of performing complex tasks. Today’s agents can search the web and pull information about electronic parts from product descriptions, images and documents.
“Some of our previous research made me curious about how LCA experts perform environmental assessments — and whether that process could be automated,” said lead author Zhihan Zhang, a UW doctoral student in the Allen School. “So we interviewed LCA experts to understand the bottlenecks firsthand, and then built a system that emulates these interactions with two AI agents. Each of them mimics different roles in the LCA process.”
One agent acts as a sort of analyst, defining what information needs to be gathered and how it will fit together. It also reviews results for accuracy. The second agent is more like an engineer. It scrapes publicly available data for information on an electronic device’s components. That might entail sifting through spreadsheets, or looking up images of the insides of devices and taking chip information from them — including from sources not typically used for LCAs, such as FCC databases and posts on iFixit.
The two agents work in a loop. The first sets the scope, the second gathers information. The first then looks that information over and might send the second agent searching again, and so on. The agents then reference LCA databases to convert the complete list of parts to carbon estimates.
The team also developed a new method to bypass this detailed data collection and directly estimate carbon footprints. For common devices like laptops and smartphones with publicly available carbon footprint reports, they found that products with similar specs like screen size and processors clustered around similar carbon values, because only a handful of companies make specialized parts for all these devices. So an unknown device’s footprint can be represented as a weighted average of similar products.
They also use this to estimate the carbon for materials not in LCA databases. For example, a new type of sustainable plastic could be estimated based on plastics with similar properties and chemistry.
“We tried this ‘nearest-neighbors’ approach and found that for materials, it’s actually better than the standard approach of a human picking the single closest entry,” said Zhang. “When estimating missing emissions factors in a test, the average error for our method was 23%. Human experts had an average error of 143%.”
The authors note that while the aim of the system is to help reduce carbon emissions overall, running AI models requires energy, so they’ve taken several steps to mitigate its impact. They use small AI models that aren’t as energy-intensive as general-purpose models. They also start the process by running a search to see if the device’s estimated emissions have already been calculated. If so, it can stop there. If the system does need to call its AI models repeatedly, estimating a device’s carbon footprint is currently on par with the emissions generated by brewing a cup of tea.
The team plans to collaborate with companies in the future to help automate their workflows.
“A lot of big companies have sustainability teams that perform these LCAs,” Iyer said. “Our hope is that automating this will actually free up their time, so they can spend their time reducing the carbon footprint of the products themselves, instead of hunting down elusive stats.”
Co-authors include Alexander Metzger, a UW student in the Allen School;, Felix Hähnlein, a UW postdoctoral researcher in the Allen School; Zachary Englhardt, a UW doctoral student in the Allen School; Shwetak Patel, a UW professor in the Allen School; Yuxuan Mei of Wesleyan University, who completed this research as a UW doctoral student in the Allen School; Tingyu Cheng of the University of Notre Dame; Gregory D. Abowd of Northeastern University; and Adriana Schulz of Brown University, who completed this research as a UW assistant professor in the Allen School.
This research was funded by Amazon Research Awards and the National Science Foundation. Zhang was supported by the Google PhD Fellowship.
For more information, contact Iyer at vsiyer@uw.edu and Zhang at zzhihan@cs.washington.edu.
Read on the original site
Open the publisher's page for the full experience