- US Department of Energy scientists explore concept of an ‘exocortex’
- It is presented as an extension of the scientist’s brain with word-based HCI.
- This is uncharted territory as one team member says.
AI is already a huge part of our lives, shaping the healthcare, transportation, education, customer service, retail and creative industries, to name just a few.
Researchers at the U.S. Department of Energy’s Brookhaven National Laboratory are currently exploring the concept of a scientific “exocortex,” the integration of artificial intelligence with human cognitive abilities, seen as an extension of the scientist’s brain.
The idea is to increase the cognitive abilities of researchers by bridging the human mind and a network of AI agents.
Kevin Yager, head of the electronic nanomaterials group at the Center for Functional Nanomaterials (CFN), detailed the idea in Digital discovery. The proposed exocortex is certainly ambitious in its scope, aiming to streamline specific research tasks through a swarm of communicating agents.
Each AI agent would specialize in particular tasks – whether managing scientific literature, orchestrating experiments or synthesizing data – and their collective behavior could provide a synergy that would significantly extend the cognitive reach of human scientists.
The exocortex might also facilitate scientific inspiration and imagination by relying on hallucinations.
As Yager explains: “Although hallucinations are generally undesirable, their existence is intrinsic and there is a trade-off between hallucinations and creativity. In other words, a certain amount of hallucinations is desirable to improve creativity and communication. More broadly, assessments of LLM creativity suggest that they can generate outcomes that are not trivially novel and useful to humans.
Of course, this vision is not without challenges and will require further advancements in AI, particularly in terms of reliability and efficiency in highly technical areas.
Yager says: “Research is needed to determine how best to leverage LLMs to generate agentic modules capable of performing tasks autonomously (on short deadlines) by iterating over a problem. »
The development of a seamless, word-based human-machine interface (HCI) will also be necessary to make these interactions feel like natural extensions of the human thought process.
Although still in the speculative stage, the concept of an exocortex seems to be a natural evolution for AI.
As Yager puts it, “the interaction between a swarm of AI agents – each tasked with intelligently managing access to a set of research capabilities – and a human researcher should lead to the emergence of enhanced human capabilities.” By extending the researcher’s intelligence to the exocortex, the researcher can accomplish more because he or she is able to intuitively and seamlessly integrate a myriad of physical, computational, and cognitive systems into his or her intellectual work.
In an article on Xplore TechnologyWritten by some of his colleagues, Yager issues a call to arms, stating: “We are entering uncharted territory with enormous potential benefits for nanoscience and beyond.” But no one can do it alone. We need community.
