Hey everyone, welcome back to My Weird Prompts. I am Corn, and I am here in our Jerusalem home with my brother, as always. It is a beautiful February afternoon here in twenty twenty-six, and the light coming through the window is just perfect for the deep dive we have planned today.
Herman Poppleberry here. And I have to say, Corn, I have been looking forward to this specific conversation since we wrapped up our last recording. There is just so much to unpack here, especially given how much the landscape of intelligence and artificial intelligence has shifted even in the last two years.
It is definitely a heavy one, but also incredibly fascinating. Today's prompt comes from Daniel, and it is about a topic that hits very close to home for us here in Israel, but has massive global implications. He wants us to discuss the use of neurodivergent individuals, specifically those on the autism spectrum, in specialized military and intelligence programs. He specifically pointed us toward the Israel Defense Forces and their units like Roim Rachok and Unit ninety-nine hundred.
Right, and it is such a timely topic because it sits at this very strange, high-stakes intersection of national security, the rapidly evolving world of multi-modal artificial intelligence, and our changing societal understanding of neurodiversity. It is not just a story about the military; it is a story about how we value different types of human minds in a world that is becoming increasingly automated. We are moving away from the era where we just wanted everyone to be a generalist, and we are entering this era of the hyper-specialist.
Exactly. And I think for those who are not familiar with the Israeli military structure, we should probably start by framing what these units actually do. When we talk about Unit ninety-nine hundred or Roim Rachok, what are we actually looking at? Because on the surface, it sounds like something out of a science fiction novel, but it is very much a reality that has been functioning for over a decade now.
It really is. So, Unit ninety-nine hundred is the visual intelligence wing of the Israel Defense Forces, or the I D F. Their whole world is imagery. We are talking about satellite photos, aerial reconnaissance from drones, and even ground-level sensor data. Their job is to look at the world from above and find things that should not be there, or find things that have changed since the last time they looked. Now, Roim Rachok, which translates from Hebrew as Looking Ahead or Seeing Far, is a specialized program within that framework. It was founded about thirteen years ago, back in twenty thirteen, specifically to integrate individuals on the autism spectrum into these highly technical roles. It was actually started by two veterans of the intelligence community, Leora Sali and Tamir Pardo, who realized that the unique cognitive profiles of people on the spectrum were being completely overlooked by the traditional draft process.
And the reason for that choice is not just about being inclusive or checking a diversity box, right? There is a very specific cognitive profile they are looking for. I remember reading that the founders noticed that certain tasks in visual intelligence require a level of focus and a specific type of pattern recognition that can be quite rare in the general population. It is almost like they identified a natural resource that was being wasted.
That is exactly how they framed it. We often talk about the strengths of the autistic brain in terms of attention to detail and a high capacity for repetitive tasks, but it goes much deeper than that. There is a concept in psychology called systemizing. It is the drive to analyze or construct systems. People on the spectrum often score very high on the systemizing scale. When you apply that to a satellite image that is essentially a massive, complex system of pixels, roads, buildings, and terrain, that high systemizing drive becomes a massive asset. They can spot a tiny change, like a new pile of dirt near a fence or a slightly different shadow cast by a vehicle, that a neurotypical analyst might miss because their brain is naturally trying to generalize the image to save energy.
That is a great point about the brain generalizing. Most of us are hardwired to see the forest and ignore the individual leaves unless they are pointed out to us. It is an evolutionary shortcut. If you see a predator in the bushes, you do not need to know how many leaves are on the branch; you just need to know there is a tiger. But in intelligence, the leaf is often the only thing that matters. But Herman, this brings us to the first big question Daniel raised. We are in February of twenty twenty-six. We have incredibly sophisticated artificial intelligence. We have computer vision that can scan thousands of images a second. Why are we still talking about human analysts, neurodivergent or otherwise, doing this work? Hasn't the A I caught up?
It is the million-dollar question, Corn. And it is something I have been researching quite a bit for this episode. You would think that A I would have totally automated this by now. And to be fair, A I does do a lot of the heavy lifting. It can flag anomalies. It can tell you that a building that was there yesterday is gone today. But A I has a major weakness that we are still struggling with in twenty twenty-six: it is only as good as its training data. A I looks for patterns it has been told to look for. Humans, and particularly those with the cognitive profile we are discussing, are excellent at finding things that do not fit any known pattern.
So it is the difference between a known unknown and an unknown unknown?
Exactly. If a military group is trying to hide something, they are going to use camouflage or tactics designed to fool standard algorithms. They know what the A I is looking for because they have access to similar A I. But a human analyst who has spent eight hours a day looking at the same square kilometer of desert for months? They develop a visceral sense of what that place is supposed to look like. They do not just see pixels; they see the story of that landscape. If a rock has been moved three inches, the A I might dismiss it as a shadow glitch or sensor noise. The human analyst sees it and thinks, why would that rock move? And that leads to the discovery of a tunnel entrance or a hidden sensor. It is that intuitive leap that A I still struggles with.
It is almost like they become the biological version of a high-resolution sensor. I find that fascinating. But I also want to push on that a bit. Is there a risk that we are essentially turning people into human computers? You mentioned that these roles involve looking at the same images for hours. That sounds incredibly taxing. Is the high capacity for repetitive tasks actually a gift, or is it something that we are exploiting because other people would burn out in ten minutes?
That is the heart of the ethical debate, and it is a really sharp observation. Critics of these programs often use the term pigeonholing. The argument is that by identifying a specific strength related to a person's neurodivergence and then placing them in a role that only utilizes that one strength, we might be limiting their growth as individuals. We are saying, you are the imagery guy, so stay in this room and look at images. It can feel a bit like a gilded cage. If we decide that an autistic person is only good for one thing, we might stop looking for all the other things they could be.
Right. And if the environment is strictly military, do they have the same opportunities for career advancement or to move into leadership roles? Or are they seen as specialized tools rather than developing professionals? In a hierarchy like the I D F, rank and social navigation are everything. If you are in a unit where you are valued for your isolation and focus, does that hinder your ability to climb the ladder?
In the early days of Roim Rachok, that was a major concern. But the program has evolved significantly. It is not just a three-year stint in the army. It includes a significant amount of life-skills training and professional development. They actually spend three months before they even start their service at Ono Academic College. They learn how to navigate a workplace, how to communicate with commanders, and how to manage the sensory environment of a busy intelligence base. And after their service, there is a huge emphasis on transitioning those skills into the private sector. Companies like Intel, Microsoft, and Check Point are very keen to hire veterans of these units because those skills in anomaly detection and data analysis are just as valuable in cybersecurity or software testing.
So, in that sense, it is a bridge to a career that might otherwise have been inaccessible. I mean, we know that historically, unemployment rates for adults on the autism spectrum have been tragically high, sometimes cited as high as eighty percent, even for those with advanced degrees. If the military service provides a credential that says, this person is an elite analyst, that is a powerful piece of paper to have in a country like Israel.
It really is. In Israel, army service is such a massive part of the social fabric and the resume-building process. Being able to say you were in Unit ninety-nine hundred is like saying you went to Harvard or M I T. For a group of people who historically might have been exempted from service or given very menial roles, this is a massive shift in status. It is moving from being seen as a liability to being seen as an elite asset. It changes the way their families see them, the way their peers see them, and most importantly, the way they see themselves.
I want to look at the global context for a second. Is this just an Israeli phenomenon because of our unique security situation and the universal draft, or are we seeing this elsewhere? Because I know other countries are watching this model very closely.
It is definitely spreading. The United Kingdom's Government Communications Headquarters, or G C H Q, has been very open about the fact that they actively recruit neurodivergent individuals. They have even said that they have a higher proportion of dyslexic and autistic staff than the general population. They see it as a cognitive diversity requirement. If everyone in your intelligence agency thinks the same way, you are going to have massive blind spots. You need the person who looks at a problem sideways. They actually have a neurodiversity support group within G C H Q that helps with workplace adjustments.
That makes total sense. In the United States, I know the National Geospatial-Intelligence Agency, the N G A, has looked into similar initiatives. And even in the private sector, as you mentioned, there is this movement called Neurodiversity at Work. It is the idea that we should not just accommodate neurodivergence, but actively seek it out as a competitive advantage. I think of SAP's Autism at Work program, which started years ago and has been a model for many other Fortune five hundred companies.
Exactly. But we have to be careful with the language we use. When we say competitive advantage, we are back to that question of exploitation. Are we valuing the person, or are we just valuing the output of their specific cognitive quirk? I was reading a paper by a neurodiversity advocate who argued that we need to be careful not to create a hierarchy of autism, where the people who can do high-level coding or imagery analysis are the good autistic people, and everyone else is still left behind. We cannot let the business case for neurodiversity become the only reason we accept neurodivergent people.
That is a really important distinction. It is the savant myth, right? The idea that every autistic person has a secret superpower like Rain Man. In reality, the spectrum is vast, and many people have challenges that do not neatly translate into a high-paying tech job or an intelligence role. If we only celebrate the ones who can serve the state or the corporation, we are still missing the point of true inclusion. We are just creating a new kind of meritocracy that is still exclusionary.
Absolutely. And even within these units, it is not always a smooth ride. The sensory environment is a huge factor. Imagine an intelligence room with dozens of screens, humming servers, bright fluorescent lights, and people constantly talking on headsets. For someone with sensory processing sensitivities, that could be a literal nightmare. One of the things Roim Rachok does well is that they actually modify the environment. They might have quieter rooms, adjustable lighting, or allow the analysts to use noise-canceling headphones. It is a level of accommodation that you do not usually see in a military setting, which is traditionally about breaking the individual to fit the system.
It is a bit of a paradox, isn't it? The military is usually the ultimate place for conformity and breaking down the individual. Yet here, they are being forced to adapt to the specific needs of the individual to get the best out of them. It is a very pragmatic form of empathy. They are realizing that if they want the talent, they have to change the culture.
Pragmatic empathy. I like that phrase. It really describes the Israeli approach to a lot of things. We have a problem, we have a resource, how do we make them fit? But let's go back to the A I piece for a minute, because I think there is a deeper layer there that is relevant to twenty twenty-six. There is this idea that as A I gets better at the things humans are currently doing in these units, the role of the human analyst will shift from finding the needle in the haystack to explaining why the needle matters.
Right, the contextualization. A I can find a pattern, but it does not understand intent. It can tell you that a convoy is moving toward a border, but it does not know if that is a scheduled exercise or the beginning of an invasion unless it has been given a very specific set of rules. And even then, it lacks the cultural and historical nuance that a human has.
Exactly. And this is where the specific cognitive style of some neurodivergent individuals becomes interesting. There is a theory called Weak Central Coherence. It suggests that many people on the spectrum focus on local details rather than the global whole. While that is often framed as a deficit in social situations, in intelligence, it means you are less likely to be biased by the big picture. You are not going to ignore a detail because it does not fit the narrative you have been told. You see the detail first, and you let the details build the narrative.
So it is a bottom-up approach to thinking versus a top-down approach. Most of us are top-down. We have a theory, and we look for evidence to support it. These analysts are bottom-up. They gather the evidence, and the theory emerges from the data. That is incredibly powerful for avoiding intelligence failures where everyone just assumes something is true because it has always been true. It is the ultimate antidote to groupthink.
It really is. If you have a room full of people who are naturally inclined to challenge the consensus simply because they are looking at different data points, you are much more likely to catch the outlier. But here is the friction point. In a military hierarchy, challenging the consensus is not always rewarded. If a twenty-year-old analyst tells a colonel that his tactical map is wrong because a bush has moved, that creates a social tension that the military is not always equipped to handle.
I can only imagine. The social nuances of military rank combined with the direct, often blunt communication style of some autistic individuals? That sounds like a recipe for some very interesting debriefing sessions. How does the I D F handle that?
They actually have to train the commanders just as much as they train the analysts. They have to tell the officers, look, when this person tells you you are wrong, they are not being disrespectful. They are being accurate. And in this room, accuracy is the only rank that matters. It is a total subversion of traditional military culture. They use what they call shadows or job coaches who act as a liaison between the analyst and the commander to help translate those social cues.
So, let's talk about the empowerment versus pigeonholing debate again. If you were an autistic young adult in Israel today, would you see this as a dream opportunity or a trap?
I think for many, it is genuinely empowering. For the first time in their lives, they are being told that the way their brain works is not a bug, it is a feature. That is a massive psychological shift. They are given a clear path to contribute to their country, a peer group that understands them, and a set of skills that the market values. But, and this is a big but, it only works if they have a choice. If the system decides for you that because you are autistic, you must go into Unit ninety-nine hundred, then we are back to pigeonholing. True empowerment requires agency. In Israel, the draft is mandatory, so the element of choice is already complicated, but Roim Rachok is a volunteer program within that framework.
And that is the challenge with any specialized program. It can become a track that is hard to get off. I wonder if we will see this model expand into other areas. Like, why stop at visual intelligence? What about code breaking? What about financial auditing? What about looking at medical imaging for rare cancers? We are seeing the rise of neurodivergent-led startups in the biotech space for exactly this reason.
Oh, it is already happening. There are companies that specialize in hiring neurodivergent teams for software testing because they are so much better at finding the edge cases, the weird little bugs that only happen when you click a button three times while a file is saving. Most people would never think to test that. But a systemizing brain will go through every possible permutation until it breaks. It is a form of high-tech artisanship.
It is like we are finally starting to map the geography of human cognition and realizing that there are all these specialized regions that we have been ignoring because they did not fit the standard office worker mold. We spent the twentieth century trying to make everyone into a standard office worker. We wanted people who were fairly good at everything and could sit in a cubicle for eight hours.
That is exactly it. In the twenty-first century, with A I taking over the fairly good at everything tasks, we need the people who are exceptionally good at one specific thing. The specialists. The outliers. The people who can do what the A I cannot. But I do worry about the downstream consequences. If we start designing roles specifically for neurodivergent brains, do we end up creating a more fragmented society? Do we lose that shared middle ground?
That is a deep concern. If we become a society of hyper-specialized silos, how do we talk to each other? But I would argue that we are already fragmented. We just used to hide it by forcing everyone to pretend they were neurotypical. All we are doing now is bringing that diversity into the light and trying to find a way to make it productive. It is about moving from a model of assimilation to a model of integration.
I want to circle back to something Daniel mentioned in his prompt, which was the idea of A I assisting these analysts. We talked about how A I is not perfect, but how does the collaboration actually look in twenty twenty-six? Because I think that is where the real future lies. It is not A I versus human; it is the human-A I hybrid.
In Unit ninety-nine hundred, they use a system where the A I does the initial scan. It might process ten thousand images and flag five hundred as having some kind of change. Then the human analyst goes through those five hundred. But here is the cool part: as the analyst makes decisions, the A I learns from them. If the analyst says, no, that is just a change in soil moisture because of the rain, the A I gets better at recognizing rain patterns. The analyst is essentially a teacher for the A I.
Exactly. And because these analysts have such a high degree of accuracy and a different way of looking at things, they are actually training the A I to be more like them. They are teaching the A I to see the details they see. It is a form of cognitive offloading. Eventually, the A I will be able to handle the rain patterns, and the human analyst can move on to even more complex, more subtle anomalies. It is a constant upward spiral of capability.
That is a fascinating feedback loop. But it also means that the bar for the human keeps getting higher. As the A I gets smarter, the human has to find even more obscure patterns to stay relevant. It is a race, for sure. But it is a race that plays to the strengths of the neurodivergent brain. Because if there is one thing we know, it is that the human mind, especially one that is wired differently, is incredibly deep. We are nowhere near the bottom of what a human can perceive.
I think about the ethical side of this too, especially in a military context. If you have an analyst who is incredibly good at finding targets, there is a moral weight to that. Does the program address the psychological impact of being that good at something that has lethal consequences? These analysts are often looking at real-time feeds. They are seeing things happen in the real world. For someone who might already struggle with emotional regulation or who has a very intense sense of justice, that can be incredibly heavy.
That is one of the most difficult parts of the conversation. The programs do have psychologists and social workers who are trained in neurodiversity, but it is a lot to ask of anyone, let alone someone who might process the world more intensely than others. It brings up the question of whether these individuals are being given the full picture of what their work entails. But then again, that is a question for every soldier in every intelligence unit.
It is. And I think the argument from the program's supporters is that by giving these individuals the chance to serve, you are giving them the same right to take on that moral burden as any other citizen. To exclude them from the burden is also a form of exclusion from the society. Citizenship isn't just about rights; it's about shared responsibilities, even the heavy ones. In Israel, where the army is the great equalizer, being excluded from that burden can be a lifelong social handicap.
That is a very profound way of looking at it. It is about the right to be responsible. And I think that is why these programs are so respected here. It is not just about the intelligence output. It is about the fact that we are a small country and we cannot afford to leave anyone behind. We need every brain we have. It is a survival imperative that has led to a social innovation.
So, looking forward, do you think we will see a day where we do not need these specialized units because the workplace has become so naturally inclusive that neurodivergent people just find their way into these roles anyway? Or is the military always going to be the pioneer here because the stakes are so high?
I would love to say yes, but I think the specialized training is still going to be necessary for a long time. Our education systems and our standard hiring practices are still so biased toward a very specific type of social intelligence. Until we change how we teach and how we hire across the board, these bridge programs like Roim Rachok are vital. They provide the support and the translation layer that the rest of the world is not yet ready to provide.
The translation layer. I like that. It is about translating between different cognitive styles. And the more we do it, the more we realize that the translation goes both ways. We are not just teaching neurodivergent people how to work in our world; we are learning how to work in theirs. And honestly, I think our world could use a bit more of that bottom-up, detail-oriented, systemizing clarity. We are so often blinded by our own narratives.
I could not agree more. It feels like we are just at the beginning of this journey. We are starting with military intelligence because the need is so clear, but the implications for medicine, for science, for art, for everything, are just massive. Imagine a neurodivergent team tasked with looking at climate data or genomic sequences. The patterns they might find could change everything. We are talking about cognitive diversity as a survival strategy for the species.
That is the big picture, Corn. And I think that is why Daniel's prompt is so important. It forces us to look past the headlines about military units and see the broader shift in how we understand what it means to be a productive member of society. We are moving from a world of standardization to a world of specialization.
Well, I think we have covered a lot of ground here, Herman. From the specific mechanics of Unit ninety-nine hundred to the deep ethical questions of empowerment versus exploitation. We have looked at the global context and the way A I is changing the game in twenty twenty-six.
It is a lot to chew on. And I am sure our listeners have their own thoughts on this, especially those who are neurodivergent themselves or have family members who are. This isn't just an abstract policy discussion; it is about people's lives and their place in the world.
Definitely. We would love to hear from you. This is such a personal and complex topic, and we know our perspective is just one part of it. If you have experience with these kinds of programs or thoughts on the ethics of cognitive-matching in the workplace, please reach out.
Absolutely. You can find us on all the usual platforms. And before we wrap up, I just want to say, if you are enjoying these deep dives, we would really appreciate it if you could leave us a review on your podcast app or a rating on Spotify. It genuinely helps other people find the show and keeps us going.
It really does. We love doing this, and knowing there is an engaged audience out there makes all the research worth it. You can find us on Spotify, Apple Podcasts, and pretty much everywhere else you get your podcasts. Our website is myweirdprompts dot com, where you can find our full archive and a contact form. And you can always email us at show at myweirdprompts dot com.
Thanks to Daniel for the prompt. It really pushed us into some interesting territory today. It is a reminder that the most interesting prompts are the ones that force us to look at the world from a different angle.
It really did. Alright, everyone, thanks for listening to My Weird Prompts. We will be back soon with another exploration of the strange, the technical, and the deeply human.
Until then, keep looking for those patterns.
Goodbye, everyone.
Goodbye.
