
Are you a postdoctoral researcher with a hankering to help the US government hone its brain-warfare skills? Well, Uncle Sam has just the job for you!
The Office of the Director of National Intelligence (ODNI) recently posted an opportunity online for a postdoctorate fellow who will āexamine how BNNs work and determine if they can be manipulated.ā For those of you unfamiliar with neuroscience acronyms, BNN stands for ābiological neural networkāāthe nervous system of a living creature.
The fellowship was posted by the National Intelligence Council on Zintellect, a job-listing site with openings available through ORAU, a consortium of universities with postdoctoral programs focused on research and education in the scientific community.
Specifically, the gig entails building a biological neural network that is trained to classify images and then trying to fool it to misclassify the images. In other words, building a brain and figuring out ways to manipulate it. The listing includes two possible ways to manufacture a biological neural network: āin vitro with neuron cell cultures or slicesā or āwith detailed physiological models of BNNs.ā It does note that, in this particular program at least, human testing isnāt authorized.
āAre the principles of adversarial attacks on BNNs different from those on [artificial neural networks]?ā the fellowship description states. āCan BNNs be made more robust to such attacks? Known adjacent phenomena are optical illusions, confirmation bias, and (more obtusely) camouflage.ā
With artificial neural networks, or ANNs, an adversarial attack is when an attacker intentionally causes a machine-learning model to make a mistake. This fellowship description is asking someone to find out whether living creatures are vulnerable to similar adversarial attacks. But you canāt apply the same type of attack on a biological neural network (read: a living creature) that you can on a machine. Itās not a replicable model. An adversarial attack on an ANN involves adding a subtle but disruptive deviation that throws off the machine learning model. Because ANNās inputs and outputs are numeric, while a BNN consists of living cells, the two networks donāt process images in the same way and thus cannot be manipulated to misclassify them using the same techniques.
It makes sense that the US intelligence community is trying to figure out how BNN manipulation works. Zachary Chase Lipton, an assistant professor at Carnegie Mellon University, told Gizmodo that adversarial attacks against computer vision systems are a clear intelligence or security interest. He gave the example of using facial recognition to control access to places or devices, noting that āunderstanding the vulnerability of those systems is essential.ā If technology existed that allowed someone to fool and gain access to those systems, the intelligence community likely wants to have access to that tech before āsome hostile country did.ā
If someone can figure out how to construct, train, and fool a brain, knowledge of those vulnerabilities could prove invaluable to the intelligence community. It doesnāt require mental gymnastics to understand why ODNI, which oversees 17 US intelligence agencies, might want to learn how to subtly exploit living creatures. In addition to widespread surveillance around the world, the US intelligence community deals with cybersecurity threats, weapons of mass destruction, and counterterrorism.
In a statement emailed to Gizmodo, an ODNI spokesperson explained that the Postdoctoral Research Fellowship Program āprovide[s] fellows an opportunity to use their expert knowledge to perform research on a specific topic for two-three years.ā
āThe person awarded an IC Postdoctoral Research Fellowship is not considered an employee of the U.S. government, and there is no continued employment,ā the spokesperson explained. āThe opportunity you are referring to, as well as all the other IC Postdoc Research Fellowship opportunities listed on Zintellect, were created by government scientists to inspire new, different, and innovative approaches to unclassified IC research.ā
Lipton doesnāt see this as a job for just any postdoctoral fellow. He pointed out that ābillions of dollarsā have been spent on research on how to simulate brains and that we are currently āat the level of creating the simulations of worm brains.ā An artificial BNN that comes anywhere close to a human brain is likely still a long, long way off.
āThis isnāt a first step,ā Lipton said of the task at hand. āThis is a Nobel Prize.ā
Update 6pm: ODNI disagrees with our description of the fellowship as a ājob,ā explaining to us that the position is a āPostdoctoral appointmentā and that fellows, who do work funded by ODNI for two or three years, are not considered employees of the US government. Weāve updated the article with language clarifying the nature of the position and added a statement from ODNI further explaining the gig.