UK’s DARPA-inspired ARIA announces first 8 program directors, research ambitions
You may have heard of DARPA, the US Defense Advanced Research Projects Agency, but have you heard of the UK’s nascent Advanced Research and Invention Agency (ARIA)?
Last month, ARIA announced its first eight program directors, along with their areas of interest.
Inspired by DARPA, ARIA’s first cohort of program directors are interested in a wide range of research projects, such as making edible vaccines from plants, developing next generation brain implants, and controlling the weather, among many others.
Here, we take a closer look at what the inaugural crew of ARIA program directors want to explore, as well as a bit of background into the new agency and its founding CEO, Ilan Gur, who came over from the US Advanced Research Projects Agency for Energy (ARPA-E).
With a background in food security, ARIA program director Angie Burnett wants to develop “new technologies that not only alter our understanding of the plants within our food system, but help us use plants to reshape it.”
For potential ARIA research programs, Burnett asks:
- Can we create edible vaccines made by plants?
- Can we program plants to do anything we want them to? For example – could we design a plant that created a rain-proof shelter from a single seed?
With a background in neuroscience and building computer chips ARIA program director Jacques Carolan says that “the devices we put in brains are decades behind the technologies we put in our pockets,” and that “building next-generation neurotechnologies will enable an entirely new suite of therapeutic tools to treat a variety of neurological and neuropsychiatric disorders, ultimately helping many more people living with disabilities.”
For ARIA, Carolan asks:
- What new technologies would enable high-bandwidth brain-machine interfaces for speech or motor control, so people living with disabilities could use the devices in their own homes?
- What new technologies could enable us to interface with the human nervous system in a targeted manner that doesn’t require invasive brain surgery? Could we use electromagnetics, acoustics, nanotechnologies, or gene therapies, or something else?
Coming to ARIA as co-program directors, Sarah Bohndiek and Gemma Bale both have backgrounds in health tech.
“We believe that there are emerging optical technologies at the edge of the possible, which will disrupt the current landscape,” they say.
“As co-PDs, we’ll look to accelerate these technologies, initially by exploring ideas around non-invasive optical mapping and sensing across a range of applications – from monitoring human health to climate change.”
For potential ARIA research programs, Bohndiek and Bale ask:
- What new breakthrough would be needed to create a 3D image of the entire human body at equivalent resolution to CT or MRI using optics? And could that technology be scaled up to image the whole ocean and everything in it?
- Could a single advance in adaptive optics allow us to enhance monitoring of greenhouse gases in the atmosphere as well as diseased cells in the body?
With a background in “examining how mathematical approaches could be used to guarantee reliable and trustworthy artificial intelligence,” David Dalrymple will look to structure and fund a concerted R&D effort toward explainable mathematical modelling of real-world phenomena using AI language models at ARIA.
“The first step is to create LLM-powered software tools that assist human scientists and engineers in developing and refining mathematical explanations of real-world phenomena, such as the dynamics of energy infrastructure, infectious disease, or the Earth climate system,” he says.
“If successful, these explanations would ground the semantics of a human-intelligible formal language in which to specify bounded tasks for future AI systems, so they perform safely and as intended.”
When considering potential ARIA research projects, Dalrymple asks:
- Can we design a unified modelling framework that is flexible enough and robust enough to cover everything from quantum chemistry to atmospheric dynamics?
- With AI assistance, might it become tractable to specify the task of safeguarding humanity from pandemics and climate change, formally enough so that a sufficiently powerful AI would be able to automate it?
With a background in physics and computational neuroscience, Jenny Read will look to move into the area of robotics at ARIA.
“Coming into the field with fresh eyes, I’ll look to facilitate links between fields that too rarely connect,” she says, adding, “most importantly, I believe there is enormous untapped potential for robots to help humans build a secure and prosperous future.”
For potential ARIA research programs, Read asks:
- Nanobots have great potential in medicine: delivering drugs, electrical stimulation or diagnostics exactly where they are needed in the body. What would take this from in vitro to in vivo?
- Insects display many complex abilities — stereoscopic depth perception, flight control, metacognition, counting — that are challenging for robots and/or were previously believed to require large brains. How are insects achieving such behavior, and what can roboticists learn from them?
With a background in electrochemistry, Mark Symes will look to “explore technologies for actively reducing atmospheric carbon dioxide levels and for climate intervention at the regional and global scale.”
We need to know what our options are for responsible climate intervention technologies, and if, where and when we should deploy them,” he says.
“If we are serious about combating climate change, we need to evaluate these technologies now.”
For ARIA, Symes asks:
- Can we develop the capability to control the weather and climate on a regional and global basis, to mitigate or obviate hurricanes, droughts, floods and heatwaves?
- What are the options for actively cooling the Earth, and how can we improve our monitor-measure-predict feedback loops to identify the most responsible choices?
With a background in advanced computing, Suraj Bramhavar has “always been enamored with how transformative technologies originate, gather momentum, and ultimately make their mark on the world.”
“The dramatic rise and power of AI algorithms coupled with the near end of exponentially improving compute performance will have massive ramifications to our society, economy, and planet,” he says, adding, “as an ARIA program director, I’ll work to create alternative hardware paradigms that can allow us to sustainably scale AI compute to benefit everyone in society.”
For ARIA, Bramhavar asks:
- What if, in trying to engineer computers that mimic the natural world in new ways, we could gain a better understanding of how the natural world processes information?
- What new ideas could emerge by breaking down the barriers between scientists on the front lines of AI algorithmic research and those building the hardware on which those algorithms run? How could these ideas impact what we know about biological function?
Potential ARIA research programs include:
- Harnessing the thermodynamics of natural systems to radically increase efficiency and scalability of AI compute.
- Imagining new ways to interface with the human nervous system to non-invasively provide neuropsychiatric treatment
- Designing programmable plants to enhance resilience against our increasingly uncertain climate.
- Catalyzing breakthroughs in advanced optics to track diseases in our bodies as well as biodiversity threats to our oceans.
The founding CEO of ARIA is former US ARPA-E director Ilan Gur, who last month said that ARIA was a DARPA-inspired startup funding R&D.
Previously, Gur founded Activate.org, a US-based organization that empowers scientists and engineers to bring ground-breaking research to market.
He is also a Schmidt Futures Innovation Fellow.
ARIA is a non-departmental public body, sponsored by the UK Department for Science, Innovation and Technology.
According to the British government, “ARIA has taken inspiration from the successful US-based Advanced Research Projects Agency (ARPA), which was instrumental in creating transformational technologies such as the internet and GPS, changing the way people live and work, while increasing productivity and economic growth.
“More recently, ARPA’s successor, DARPA, was a vital pre-pandemic funder of mRNA vaccines and antibody therapies, leading to critical COVID-19 therapies.”
ARIA bills itself as “an R&D funding agency built to unlock scientific and technological breakthroughs which could benefit everyone” by “empowering scientists to reach for the edge of the possible.”