The following article is an original work published by the Information Professionals Association. Opinions expressed by authors are their own, and do not necessarily reflect the views of or endorsement by the Information Professionals Association.
By Sean A. Guillory & John T. Carrola
Some days I take a break from video games to walk outside and look around. And then I wonder “what kind of GPU are they using to run the simulation of us because wow these graphics are amazing?” – Dr. Patrick Biltgen, Booz Allen Hamilton Principal and Citizen of a Complex World
Do you believe you can tell the difference between physical reality, virtual/information-based reality, and thoughts/imaginings in your own head? Those three differentiations are what the DoD’s Joint Concept for Operating in the Information Environment define as the three dimensions of the information environment: physical, informational, and cognitive dimension. These dimensions can permeate across the different domains (e.g. land, sea, air, etc.) and while they can overlap (e.g. cyber-physical systems, human-machine teaming, etc.), many things that we take for granted in understanding “reality” depend on people being able to differentiate between these dimensions. Things like identifying mis/disinformation, Deepfakes, or knowing that your life isn’t really in danger when playing a video game, are absolutely dependent on knowing what did or did not happen within different dimensions (e.g. you may die in a video game or in a dream but that is different than your physically dimensioned life).
So what happens when one can’t tell the difference between the dimensions anymore (even if a person thinks they can)? It may sound ridiculous but this kind of convergence isn’t just possible, it’s already happened to many people. The technological aims of the next ten years threaten to affect it for the majority of the population. Examples include Metaverse/Web3, synthetic training environments, Integrated Visual Augmentation Systems, digital twins, brain-machine interfaces and other biodigital convergences, technology to make video-recorded people say or do whatever one would like, and even capabilities that can link up all of the above. We have coined the term “Online-Offline (O-O) Convergence” to describe when a person can’t differentiate between the information environment dimensions and sees it as one “reality.” Some researchers have dubbed this converged dimension “Onlife” but we’ll leave it nameless for now.
As mentioned in the first paragraph, the ability to differentiate the dimensions is needed to identify things like mis/disinformation (i.e. being able to discern if purported information derived from either the physical or information environment and know which parts have been fabricated or misremembered). However, if the dimensions cannot be easily discerned by the perceiver of the experience (O-O Convergence), then discerning what’s mis/disinformation becomes impossible since what one experiences is a single dimensional reality (i.e. all of the information a person takes in would be seen as equally valid because it all comes off as equally “real”). Instead of trying to assess the validity of the information, assessment would be on if a person actually had a certain experience or not (which we’ll coin “misexperiences/disexperiences”). This is a concern because in an O-O converged environment, experiences can be individually tailored where it’s possible for everyone to be seeing and hearing different things at the “same” event. Examples of this could be purposely enacted by the users to have “experience filters” that would be like putting the equivalent of Snapchat Filters onto an entire experience (e.g. a filter to make everyone look like Simpsons characters, manipulate the experience to make users feel like a “Giga-Chad”, or gives the experience of “what a serial killer would see”) or these same sorts of experience filters can be enacted by the platform without the user knowing (e.g. an event where certain people’s accents and ethnicities are changed based on what the algorithm infers to be the preference of the experiencer). It’s also extremely important to note that while this paper has mainly talked about the experiences of individuals, the importance is that these individual personalizations can be done at across-platform scale (a point also mentioned by Rand Waltzman in the 100th episode of The Cognitive Crucible) to help create, curate, and manipulate groupings of people to essentially live in different “realities” (not just consuming different information like today’s information environment; consuming different realities).
Please don’t think that simply knowing about the O-O Convergence is enough to negate it (the G.I. Joe catchphrase “knowing is half the battle” doesn’t work with cognitive biases); please don’t assume that people who “fall” for these things are either “dumb” or mentally impaired; and please do not think that O-O Convergence will only be a blue-collar/rank-and-file populace-only issue. Imagine what a world would look like where it is truly difficult to discern if information is “real”/reliable (be it printed, online, or even one’s memories); the foundations of policy, law enforcement, science, and our democracy rest on being able to discern what is “real” or not across the information environment. Many worry about a “Singularity” in which computer abilities will overtake the ability of the human brain (which the author Ray Kurzweil predicts will be 2045); we would argue that an even bigger issue – when most humans can’t differentiate between the three dimensions (i.e. The O-O Convergence Point) – is coming sooner than 2045 (which notes that the Singularity assumes people are able to tell the difference between the physical, information, and cognitive dimensions in order to be able to assess the computer vs. human intelligence).
And how close are we to this O-O Convergence Point? With the current conflict in Ukraine being described as the first “TikTok War,” it certainly seems close. While kinetic battles are being fought in the physical dimension, those experiences are being mirrored in the information dimension where battles for attention and influence are fought, and in the cognitive dimension where decisions on who is winning or losing are contextualized to the backdrop of things like norms, consensus, archetypes, and historical knowledge. The flow of that experience moves faster than ever but there are still choke points that demonstrate the dimensions still have some separation (e.g. the time it takes to edit and upload a TikTok; the hashtagging/search engine optimizing/@-ing it takes to make sure a message gets noticed by the right audience, etc.). Even just having the integration of a few technologies (e.g. having AR/XR with automated captioning/labeling for physical/information dimensional integration) can make those things run seemingly seamlessly where it appears as one continuous, inseparable dimension to many people. People like to talk about the lead up to the 2016 U.S. Presidential election as a “meme war,” but in the past few months of the first TikTok war, we’ve had patron saints for javelin missiles, ghosts of Kyiv being treated as “real enough” in the face of other evidence, Russian soldiers and UN officials considering an actress of a Ukrainian propaganda video as “no longer human,” and many other examples of online memes actually influencing the kinetic warzone in a “national will to fight” that was underestimated both by foe and allies alike. Online memes have tasted human blood and with the technology integrations discussed above moving more to the commercial sector within the next 10 years, one should expect these future “meme wars” to be a lot more deadly the closer we move towards the O-O Convergence Point.
Let us reiterate that we do believe that there are going to be many positives that come from these technologies that help in the convergence of the different dimensions, and for some important sensor/intelligence/information fusion initiatives within Defense, they will be critical for our nation and allies keeping our advantage (e.g. for things like JADC2 to be a reality where users don’t immediately succumb to “cognitive overload,” there would need to be “Cognitive Mission Support” systems [term coined by the brilliant Ki Lee] that would enable O-O Convergence to help ease the cognitive load to engage with all of the information). We should continue aiming for those positives; however, we must avoid the same “irrational exuberance” that had many people thinking the advent of social media would lead to some sort of “Golden Age” for discourse and societies – only to be blinded in how such technology could be weaponized. Note that with our present technology we have people storming a pizza place looking for a basement full of captured children, lynching people for believing a staged training video was real, livestreaming one’s mass shootings while mentioning memes and YouTube stars in-between reloadings, or even committing a genocide over misinformation. This is all happening now when still many (I hesitate to say “most”) people can still discern between the dimensions: the Deepfakes keep getting more “real” by the day; AI is getting so good it’s convincing its own creators that it is conscious (with a jury of our peers deciding if that’s the case being likely to happen within the next 10 years); and spree killers are going for “high scores” already. What of a reality where terms like “gamification” are meaningless because it’s impossible to tell the difference between video games and “reality?” And if there is any doubt that there aren’t folks who are trying to enable and weaponize the aspects of O-O Convergence, we leave you with this extended quote from a recent interview with Steve Bannon from The Atlantic:
“[Bannon] notes how stunned he was to discover how many people played multiplayer online games, and how intensely they played them. But then he breaks it down for [Errol] Morris, using the example of a theoretical man named Dave in Accounts Payable who one day drops dead.
“Some preacher from a church or some guy from a funeral home who’s never met him does a 10-minute eulogy, says a few prayers,” Bannon says. “And that’s Dave.”
But that’s offline Dave. Online Dave is a whole other story. “Dave in the game is Ajax,” Bannon continues. “And Ajax is, like, the man.” Ajax gets a caisson when he dies and is carried off to a raging funeral pyre. The rival group comes out and attacks. “There’s literally thousands of people there,” Bannon says. “People are home playing the game, and guys are not going to work. And women are not going to work. Because it’s Ajax.”
“Now, who’s more real?” Bannon asks. Dave in Accounting? Or Ajax?
Ajax, Bannon realized. Some people—particularly disaffected men—actively prefer and better identify with the online versions of themselves. He kept this top of mind when he took over Breitbart News in 2012 and decided to build out the comments section. “This became more of a community than the city they live in, the town they live in, the old bowling league,” he tells Morris. “The key to these sites was the comment section. This could be weaponized at some point in time. The angry voices, properly directed, have latent political power.”
I mentioned this moment to Bannon the second time we spoke. On War Room, he frequently talks about three levels of participation: the posse, the cadre, and the vanguard. It sounded to me like the gamification of politics. Yes, he told me. That’s just it: “I want Dave in Accounting to be Ajax in his life.”
But that’s precisely what happened on January 6. The angry, howling hordes arrived as real-life avatars, cosplaying the role of rebels in face paint and fur. They stormed the Capitol while an enemy army tried to beat them away. They carried their own versions of caissons. They skipped a day of work. And then they expressed outrage—and utter incredulity—when they got carted away.
The fantasy and the reality had become one and the same.”
Remember, that’s what we can do today; imagine when we get to the point where the majority of people wouldn’t be able to differentiate what is “Dave” and “Ajax” (making the online-offline weaponizations expressed above easier than ever). We’re sure the graphics will be amazing.
About the authors
Dr. Sean Guillory utilizes his cognitive neuroscience training to help with cognitive/human domain capabilities within Defense and National Security. He is also a member of the Information Professionals Association.
John Carrola, M.S., utilizes his technical expertise of deep-learning model evaluation and optimization to help with various information environment capabilities within Defense and National Security.