Georgia Tech has switched on a high-tech cloud laboratory in Atlanta that lets researchers anywhere in the country submit experiments virtually and have robots carry them out, part of a push that could shrink years of manufacturing and materials research down to a matter of weeks. The lab uses pre-programmed and programmable processes on demand, with cloud laboratory experts running the submitted experiments and sending results back to researchers who never have to set foot on campus.
The project is backed by an $18 million grant from the National Science Foundation, according to 95.5 WSB, which reported that Jonathan O'Brien contributed to the story. Georgia Tech projects the facility will serve more than 400 researchers across 150 academic, industry, and government organizations nationwide, with over half of them participating remotely, according to WSB-TV.
The goal, per the WSB report, is to reduce research costs from what can run tens of millions of dollars down to tens of thousands, while cutting a research process that traditionally takes six to ten years down to a few weeks or months. Specialized equipment training alone can eat up three, four, or five weeks per piece of work under the old model — time researchers can now skip since the cloud lab handles the hands-on work remotely.
From Pilot Facility to National Resource
The cloud lab is built on top of Georgia Tech's existing Advanced Manufacturing Pilot Facility, which already runs autonomous workflows across roughly 38 pieces of equipment. Under the new grant, the school plans to expand automated workflows to more than 100 of the site's 160 total machines, according to the Georgia Tech Research News Center.
Duke University is contributing computational firepower to the effort, integrating its aflow.org materials discovery database — which holds data on more than 3.5 million material compounds and 730 million calculated properties — so virtual predictions can be paired with physical robotic testing, according to the Duke Pratt School of Engineering. The lab is designed to advance manufacturing, electronics, and medicine by allowing researchers to work remotely, refine experiments, and access advanced manufacturing capabilities they might not have on their own campuses.
The artificial intelligence steering the lab's autonomous decision-making comes from the Georgia Tech-based NSF Artificial Intelligence Institute for Advances in Optimization, led by Professor Pascal Van Hentenryck, which is supplying the algorithms and optimization frameworks behind the robotic workflows, per AI4OPT. Aaron Stebner, associate director of the Georgia Tech Manufacturing Institute, said in August that Georgia Tech had been working toward this kind of AI and advanced manufacturing integration for six years — reaching a benchmark he originally figured would take two decades, according to the Georgia Tech Research News Center.
Part of a Bigger Federal Bet
Georgia Tech's award was one piece of a broader federal rollout announced July 22, when the National Science Foundation established a nationwide network of 20 AI-enabled cloud laboratories designed to interoperate and share research workflows, per the Georgia Tech Research News Center. Access to the lab is currently restricted to researchers, though Georgia Tech has said it plans to expand access beyond academic use — with hobbyists working in garages among those who could eventually benefit, and the broader network intended to grow into a national resource.
Other sites in the same NSF network are tackling different scientific domains. The California Institute of Technology received $17 million to build ELECTRA, an automated electron diffraction cloud lab aimed at identifying unknown 3D molecular structures remotely, as reported by Pasadena Now. Northwestern University landed $20 million to launch the DREAM Cloud Lab, billed as the nation's first publicly accessible AI-powered cloud lab dedicated to protein engineering, according to Northwestern Now.
Atlanta's Growing AI Infrastructure
The cloud lab isn't Georgia Tech's only recent haul of federal AI money. The university also received a separate $20 million NSF award in July to build Nexus, an AI-focused supercomputer capable of more than 400 quadrillion operations per second, intended to support AI research in medicine, robotics, and materials science.
It also builds on groundwork laid earlier this year: Georgia Tech's self-driving factory lab debuted in Midtown Atlanta in May as an AI-driven pilot facility to test industrial robotics and train workers. Together with the new cloud lab and the Nexus supercomputer, the additions point to Atlanta steadily consolidating itself as a hub for federally funded AI infrastructure. How quickly Georgia Tech can automate its remaining equipment, and exactly how non-academic users will eventually gain access, remain open questions as the project ramps up.
