You know Herman, living here in Jerusalem, you get used to a certain level of background noise when it comes to news. There is always something happening, always some headline about tensions or diplomatic moves. But there is a huge difference between that constant buzz and the kind of hard, actionable information you actually need when things get serious, especially given the regional volatility we have seen through twenty twenty four and twenty twenty five.
Absolutely. It is the difference between reading a weather report that says it might rain and having a high resolution radar map that shows exactly where the storm cell is moving. Herman Poppleberry here, and I have to say, our housemate Daniel really hit on something profound with this prompt. He is looking for that precision, that SITREP style, situational report style, specifically for the security situation in Israel.
Right, he mentioned how during those intense rounds of hostility between Israel and Iran, or the multi front conflicts we have been tracking, the standard news cycle just feels so fat filled. It is mostly speculation, talking heads, and emotional commentary. Daniel is looking for the lean, high protein version of information. He mentioned the Presidents Daily Brief and the Institute for the Study of War reports as his benchmarks.
Those are the gold standards. And what is fascinating is that they are not just informative, they are a specific art form. There is a whole philosophy of writing behind them that is designed to minimize cognitive load for a decision maker while maximizing clarity and credibility. In the intelligence community, they call it tradecraft.
I love the idea of digging into the anatomy of these reports. Because if we can deconstruct what makes them good, we can figure out how to get an artificial intelligence to do it for us. It is about moving from being a passive consumer of news to being an active architect of your own intelligence brief.
Exactly. And I think we should start with the hierarchy of information. In these high level briefs, they use a concept called B L U F, or Bottom Line Up Front. Most journalism is written in an inverted pyramid style, but an intelligence brief takes that to an extreme. If the Prime Minister or the President only has thirty seconds to read a page, what is the one thing they must know? That goes at the very top, usually in a bolded top line.
It is almost like the opposite of clickbait. Instead of burying the lead to keep you scrolling, they are giving you the lead immediately so you can stop reading if you need to. But let us talk about that hierarchy. Daniel mentioned bullet points and graphics. Why are those so much more effective than a well written paragraph?
It comes down to scanability and the reduction of ambiguity. A paragraph allows for nuance, but it also allows for fluff and hedging. Bullet points force the writer to make discrete, falsifiable claims. If you have five bullets, those are five specific facts or assessments. You cannot hide a lack of information behind flowery prose. It also prevents the narrative fallacy, where a writer tries to connect dots that might not actually be connected just to make a story flow better.
That is a great point. It is harder to lie or speculate when you are forced into a bulleted list. But there is also the element of time. Daniel mentioned precise time stamping and the use of Coordinated Universal Time, or Zulu time. In a fast moving security situation, saying something happened this morning is useless. Was it six A M or ten A M? That four hour gap could mean the difference between a threat being active or neutralized.
And it prevents the echo chamber effect. You see this on social media all the time during a crisis. Someone posts a video, and it gets reshared for twelve hours. Without a precise time stamp, a ten hour old event looks like it is happening right now. A proper situational report anchors every piece of data in time and space. It also helps identify circular reporting, where three different news outlets are all just quoting the same unverified tweet.
I think another key element Daniel mentioned was source attribution. In these reports, they are very careful about where the information came from. Is it a social media report? Is it an official military statement? Is it a human intelligence source? They often use qualifiers for their confidence levels too.
Yes, the intelligence community has a very specific vocabulary for this. They will say things like, we assess with high confidence, or there is a moderate probability. They are not just giving you the news, they are giving you the metadata about the news. They are telling you how much you should trust each piece of information. High confidence usually means the information is confirmed by multiple independent sources; low confidence means it is a single source or a plausible inference.
This is where I think most people get it wrong when they try to follow a developing situation. They treat all information as having equal weight. But a situational report is essentially a weighted graph of information. The node with the most reliable sources and the most recent time stamp gets the most weight.
And let us not forget the graphics. When the Institute for the Study of War puts out a map of a conflict zone, they are not just making it look pretty. They are using visual language to convey complex spatial relationships. If you see a red polygon on a map, that represents a very specific area of control or influence. It tells you more in two seconds than a three page description of street names ever could.
It is funny you mention that, because I was thinking about our context here in Jerusalem. When there is a security incident, the local WhatsApp groups are flooded with text. Move here, stay away from there, this happened at this gate. But the moment someone drops a pin on a map or shares a clear graphic of the perimeter, the signal to noise ratio just flips. Everyone instantly understands the situation.
That is the power of the SITREP. It is about creating a shared mental model of reality. Now, the big question Daniel asked is how we can get an artificial intelligence to do this. Because right now, if you ask a standard chatbot for a news update, you get a summary of news articles, which is exactly what Daniel wants to avoid.
Right, the A I usually just regurgitates the fluff. It says, experts say this, or some believe that. It is very wishy washy. To get an A I to write like a situational report officer, we have to change the way we instruct it. We have to give it a very rigid set of constraints and a specific hierarchy to follow.
I think the first step is the source material. You cannot just let the A I browse the open web and pick whatever it finds. You have to feed it the raw data. If I were building this, I would start by scraping official government feeds, reputable news wires like Reuters, and verified O S I N T, or open source intelligence accounts, that are known for ground truth reporting.
And then you need a system prompt that defines the persona. You are not a helpful assistant, you are an intelligence analyst writing a daily brief for a high level decision maker. Your goal is brevity, accuracy, and the elimination of speculation.
I would even go further and give it a negative constraint list. Do not use adjectives like shocking, unprecedented, or alarming. Do not include quotes from analysts unless they contain a specific factual claim. Do not speculate on motives unless there is a stated manifesto or official declaration.
That would definitely cut out the fat. But what about the structure? How do we get it to handle that precise hierarchy Daniel likes?
You use what we call few shot prompting. You provide the A I with five or ten perfect examples of a situational report. You show it exactly how the top lines look, how the bullet points are nested, and how the time stamps are formatted. Artificial intelligence is incredibly good at pattern matching. If you give it a rigid enough template, it will stick to it.
I wonder if we could even instruct it on how to handle conflicting information. In a real security situation, you often have source A saying one thing and source B saying another. A typical A I might try to average them out or pick one. But a good SITREP would list both and assign a confidence level to each based on the source quality.
That is exactly what you would want. You could tell the A I, if two sources conflict, create a sub bullet for each, cite the source, and highlight the discrepancy. This is actually a second order effect of good intelligence writing. It reveals the gaps in our knowledge. Sometimes the most important part of a brief is knowing what we do not know yet.
That is a great insight. Most news tries to present a complete, closed narrative. An intelligence brief is an open system. It says, here is what we have confirmed, here is what is unconfirmed but likely, and here is the massive hole in our information regarding this specific sector.
And for someone like Daniel, or anyone living in a complex environment, knowing where the information gaps are is a survival skill. It tells you where you need to be more cautious.
Let us talk about the data for a second. Daniel mentioned he wanted this for the security situation in Israel. If we were to build this today, what are the specific sources we would point an A I toward to get that intelligence grade reporting?
Well, you would definitely want the official Israeli Defense Forces feeds, but you have to account for their specific bias and operational security delays. You would want the United Nations Office for the Coordination of Humanitarian Affairs for ground level impact reports. You would want the major wire services like the Associated Press for verified event timing. And then you would want a curated list of open source intelligence accounts, people who specialize in geolocating footage and verifying satellite imagery.
The challenge there is that the A I needs to be able to synthesize all of that in real time. We are talking about an update that might need to be issued every few hours during a crisis.
Exactly. And that is where the precise time stamping becomes a technical requirement. The A I needs to be able to sort its input data by time and prioritize the most recent information. It has to be able to say, this report from ten minutes ago supersedes the one from two hours ago, even if the older one was more detailed.
I think one thing most people miss about these high level briefs is the so what factor. Every piece of information in a SITREP should have an implication. If a road is closed, the so what is that supply lines are cut or civilian evacuation is blocked.
Right. In the intelligence world, they call that the assessment. You have the fact, and then you have the assessment of what that fact means for the overall situation. You could instruct an A I to always follow a factual bullet with an assessment bullet. Fact, the border crossing at Kerem Shalom is closed. Assessment, this will likely lead to a twenty percent reduction in fuel deliveries within the next twenty four hours.
That is where it gets really powerful. It turns raw data into actionable intelligence. But you have to be so careful with the A I there, because that is where it is most likely to start hallucinating or getting overconfident in its predictions.
True. You would need to keep the assessments very grounded in logistical and physical realities. No psychological speculation. Stick to the physics and the geography of the situation.
You know, it occurs to me that this style of writing is actually useful for almost any professional field. If you are a doctor, a lawyer, or a software engineer, you would probably much rather read a SITREP of your daily work than a long, rambling email chain.
Oh, absolutely. I have actually started doing this with my own project management. Instead of writing long updates, I use a SITREP format. Top lines for the biggest blockers, bullet points for completed tasks with time stamps, and a section for unknown risks. It has changed the way my team communicates. It cuts out all the ego and the fluff and just gets to the heart of the matter.
It is funny how a format designed for the highest stakes military and political decisions can be so refreshing in a mundane office setting. I think it is because we are all suffering from information overload. We are drowning in words, but starving for clarity.
And that is the core of Daniel's prompt. He is looking for clarity in a very high stakes environment. It makes total sense that he would be drawn to the WikiLeaks cables or the I S W reports. Those writers are trained to value the reader's time above all else.
I want to go back to something Daniel mentioned about the WikiLeaks cables. He said he was enamored by the way they presented information. It was not journalism, but it was a fantastic mix of analysis and facts in a tightly packaged format. What is it about that specific diplomatic style that is so compelling?
It is the candor. In a diplomatic cable, the writer is speaking to their own government. They are not trying to be polite or politically correct for a general audience. They are being brutally honest about what they see on the ground. They are describing the personalities of leaders, the hidden tensions in a room, the real reasons a deal fell through.
So, it is the combination of the rigid SITREP structure with the depth of high level human observation. If we could get an A I to mimic that, while keeping it grounded in verified facts, that would be the ultimate daily brief.
It really would. Imagine an A I that could not only tell you that a protest happened, but could also analyze the composition of the crowd, the specific slogans being used, and how that compares to protests from six months ago. That kind of longitudinal analysis is where these reports really shine.
Let us do a little thought experiment. Imagine we are designing the system prompt for Daniel's Israel Security SITREP A I. What are the first three lines of that prompt?
Okay, line one. You are a senior intelligence analyst for the Institute for the Study of War, tasked with providing a daily situational report on the security environment in Israel and the surrounding region. Line two. Your writing must be strictly objective, utilizing the Bottom Line Up Front format, and every factual claim must be attributed to a specific source with a confidence level. Line three. Prohibit all speculative language, emotional modifiers, and circular reporting. If a fact cannot be verified by at least two independent sources, mark it as unconfirmed.
That is a strong start. I would add a line about the hierarchy. Use a three level nested bullet system. Level one is the strategic theme, level two is the specific tactical event, and level three is the immediate assessment of that event.
Perfect. And we should probably include an instruction on how to handle geography. Use specific coordinates or neighborhood names, and always relate the event to its proximity to major infrastructure or civilian centers.
This is making me think about the broader implications of everyone having access to this kind of tool. If we can all generate our own high quality intelligence briefs, does that make us better citizens, or does it just isolate us more in our own data silos?
That is the big question. My hope is that it makes us more discerning. When you get used to the precision of a SITREP, you start to see through the fluff of standard news. You become a more critical consumer of information. You start asking, wait, where is the time stamp on this? What is the source for that claim? What is the so what here?
It raises the bar for everyone. If the average person can distinguish between a factual report and a speculative piece of commentary, then the media has to step up its game. It can no longer get away with being vague and emotional.
And in a place like Jerusalem, where the stakes are so high, that kind of media literacy is a necessity. It is not just about being informed, it is about being accurately informed.
You know, I was reading about the history of the Presidents Daily Brief recently. It started in the nineteen sixties because President Kennedy felt he was getting too much raw data and not enough synthesized intelligence after the Bay of Pigs failure. He needed a way to see the big picture without getting bogged down in the minutiae.
It is a classic problem of the modern age. We have more information than any humans in history, but we are often less clear on what is actually happening. The P D B was an attempt to solve that at the highest level of government. What Daniel is asking for is a way to solve it at the personal level.
It is a democratization of intelligence. We are all our own little Presidents now, trying to make decisions for our families and our communities based on a chaotic stream of data.
And the A I is the perfect tool for this because it does not get tired. It can monitor a thousand feeds simultaneously and look for the specific patterns we have instructed it to find. It can do the grunt work of sorting and time stamping so we can focus on the assessment.
I think there is a real opportunity here for someone to build a platform that does exactly this. Not just for security, but for any complex, fast moving situation. Imagine a SITREP for the global energy market, or for the development of quantum computing.
I would subscribe to those in a heartbeat. It would save me hours of reading through fluff pieces and marketing jargon. Just give me the top lines, the bullet points, and the confidence levels.
It is about respect for the reader's cognitive bandwidth. In a world that is constantly trying to steal our attention with outrage and drama, a situational report is a rare act of intellectual honesty. It says, here is what we know, here is why it matters, and here is where we are still in the dark.
And it acknowledges that the reader is capable of making their own decisions if they are given the right data. It is not trying to tell you how to feel. It is just trying to tell you what is.
I think that is a perfect place to wrap up the first part of our discussion. We have broken down the anatomy of the SITREP, the importance of hierarchy and precision, and how we might start to train an A I to mimic this style. But I want to dig deeper into the practicalities. If Daniel, or any of our listeners, wanted to start building this for themselves today, what are the actual tools and workflows they would use?
That is where it gets really interesting. We are moving from the theory of intelligence writing to the engineering of intelligence systems.
Before we get into that, I just want to say, if you are enjoying this deep dive into the art of information, we would really appreciate a quick review on your podcast app. It genuinely helps other people find the show and join the conversation.
It really does. Alright, let us talk about the stack. If I were building the Daniel Brief today, I would start with a tool like Python to handle the data scraping. You need something that can talk to A P I s and pull in raw text from various sources.
And for the non programmers out there, there are actually some great no code tools that can do this too. Things like Zapier or Make can be configured to watch a specific R S S feed or a social media list and push that data into a database.
Right. The key is to get all your raw data into one place. Once you have that, you can use a large language model like G P T four or Claude three to do the heavy lifting of the synthesis.
But you have to be careful about the context window. If you feed it too much data at once, it might lose the thread or start to overlook the subtle contradictions between sources.
That is where a technique called R A G, or Retrieval Augmented Generation, comes in. Instead of giving the A I everything at once, you store all your data in a vector database. Then, when you ask for the daily brief, the system searches that database for the most relevant and recent pieces of information and feeds only those to the A I.
So, it is like giving the A I a very organized filing cabinet to look through before it starts writing the report. This would solve the time stamping issue too, because you could tell the search algorithm to prioritize the most recent entries.
Exactly. And for the graphics part Daniel mentioned, we are starting to see A I tools that can generate maps and charts directly from text data. You could instruct the A I to output a set of coordinates for any mentioned events, and then use a mapping tool to plot those automatically.
That would be incredible. A fully automated, visual SITREP that updates in real time. We are getting very close to that being a reality for the average person.
We are. But it still requires that human element of oversight. You still need to be the one setting the constraints and evaluating the output. You have to be the editor in chief of your own intelligence agency.
I think that is the most important takeaway. The A I can do the work, but we have to provide the wisdom. We have to decide what matters and what the so what is for our own lives.
And that brings us back to our context here in Jerusalem. For Daniel, the so what might be whether it is safe to walk to the grocery store or if he should stay inside today. For someone else, it might be about whether to invest in a certain company or how to plan a trip. The format is the same, but the application is personal.
It is about taking back control of our information environment. Instead of being victims of the algorithm, we become the masters of our own data.
I love that. It is a very empowering way to look at technology. It is not just about making things easier, it is about making us more capable and more informed.
Well, Herman, I think we have given Daniel a lot to think about. From the philosophy of the Presidents Daily Brief to the technical stack of an A I driven SITREP.
It has been a fascinating journey. I am definitely going to be looking at my daily news feed a little differently after this. I might even start drafting my own top lines before I have my morning coffee.
Just make sure you use Zulu time, Herman. I do not want any ambiguity about when you finished your first cup.
Noted. Twelve hundred hours Zulu, coffee levels are optimal. Assessment, productivity will likely increase by fifteen percent over the next two hours.
High confidence in that assessment. Well, thanks everyone for joining us on this episode of My Weird Prompts. It has been a blast digging into this with you, Herman.
Always a pleasure, Corn. And thanks again to Daniel for such a thought provoking prompt. It is good to have housemates who keep us on our toes.
Absolutely. You can find us on Spotify and at our website, myweirdprompts dot com, where we have all our past episodes and a contact form if you want to send us your own weird prompts.
We love hearing from you guys. Until next time, stay curious and keep asking the deep questions.
Take care, everyone. Goodbye.
Goodbye.
