The investment aims to accelerate scientific progress and automate research
The US National Science Foundation (NSF) aims to invest up to $100m to support a network of programmable cloud laboratories to expand access to technology and accelerate automation of scientific discovery.
"The initiative will transform how US researchers conduct scientific experiments,” said NSF assistant director for TIP Erwin Gianchandani. “It will accelerate scientific progress by advancing AI-enabled technologies that form the backbone of the automated science revolution.”
The NSF Test Bed: Toward a Network of Programmable Cloud Laboratories (NSF PCL Test Bed) will invest in and establish AI-enabled laboratories across the US to integrate, test, evaluate and validate the capabilities of new AI-based technologies. The laboratories can be remotely accessed to run custom, user-programmed AI-led workflows.
The idea is that this network will help to bring innovative technologies into practical use during scientific experiments, by making it possible to use AI throughout every stage of laboratory experimentation.
"The idea of a national network of programmable cloud laboratories builds on NSF's longstanding legacy of transformative investments — such as NSFNET decades ago — that paved the way for the modern internet," said Gianchandani.
“This is a crucial step toward addressing the growing need to generate and interpret large volumes of high-quality experimental data in biotechnology, materials science, chemistry and other laboratory sciences.”
For example, in the pre-experiment phase, AI could help to design the best setup or predict likely outcomes, reducing trial-and-error and saving time and money. During the experiment, AI could track real-time data through sensors or imaging tools, automatically control conditions and make real-time adjustments.
AI could even help in the post-experiment phase, where researchers could use it to plan next steps or accelerate data analysis and visualisation.
The initial focus of this project will be on biotechnology and materials science.
In addition to the AI-led laboratory network, the investment will also go to education and training, by providing access to advanced laboratories in classroom settings.
