My Weird Prompts

After gallbladder removal, millions suffer from post-cholecystectomy syndrome — bloating, bile reflux, weight gain — yet no single specialist owns the condition. This episode explores whether an LLM council (multiple AI models prompted as different specialists, then synthesized) could help patients build a coherent treatment plan. We walk through two concrete architectures: role-based councils (ga...

Stanford's Smallville and Moltbook create AI agents that plan parties and form relationships — but is this social intelligence or just elaborate puppetry? We dig into the fundamental limits of text-only architectures, from missing body language to emotional forgetting curves, and ask whether embodiment and world models could bridge the gap. This episode explores what these simulations actually rev...

AI models love bullet points — but when you're generating client reports, board memos, or regulatory filings, that default style destroys credibility. In this episode, we break down why transformers favor structured output (it's not laziness, it's the attention mechanism) and compare three engineering approaches to fix it: system prompting (62% adherence), fine-tuning ($200+), and textual LoRAs (9...

Geopolitical simulations using AI agents sound like a powerful forecasting tool—until you see the bill. One run of Snowflake, an open-source platform from the CIA’s Technical Innovation Lab, can cost over $150 in API tokens. Because LLMs are probabilistic, you need to run the simulation dozens or hundreds of times for statistically valid results. In this episode, we break down exactly where the mo...

What happens when fraudsters build tech stacks that rival legitimate startups? This episode follows two firsthand accounts from a moving scam with a purpose-built quotation engine and a rental scam run through a single password manager. We trace how custom CRMs, victim-scoring algorithms, and multi-alias management systems turn deception into an operations problem — and what that means for anyone...

What if the best way to protect your real address isn't locking it down — but publishing a fake one, loudly and convincingly? This episode explores obfuscation as a privacy strategy: seeding false information into the digital ecosystem so anyone trying to trace you hits a wall of plausible decoys. We break down how to build fake nonprofit websites, phantom addresses, and decoy personas that surviv...

Large language models have a pathological addiction to bullet points — and it's costing enterprises credibility. In this episode, we break down the three reasons LLMs default to list structures (training data distribution, RLHF rewards, and attention mechanism entropy) and compare the engineering levers available to force prose-first writing. From system prompting with few-shot examples to fine-tu...

A phone number seems like the most mundane string of digits — until you feed it into a graph-based OSINT system. This episode traces the full pipeline: from proper E.164 normalization (where most investigations silently fail) through Maltego transforms that connect numbers to social profiles, emails, and carrier data. We explore the permutation strategy that turns a single identifier into a high-c...

Most people picture infantry or tanks when they think of ground forces, but nothing moves without combat engineers going first. This episode unpacks the daily reality of the role — from route clearance and breaching calculations to the psychology of being first through a breach. We explore the IDF’s conscript-driven Combat Engineering Corps, the gap between battalion engineers and Yahalom special...

Sun Tzu and Machiavelli are the most-quoted strategic authorities in business, but they fundamentally disagree on what strategy even is. This episode pulls apart their actual philosophies — from deception and cruelty to victory and survival — and shows why the lazy mashup produces genuinely bad decisions in boardrooms, startups, and even AI strategy. If you've ever heard "we need to be more Sun Tz...

In a single-party consent jurisdiction like Israel, tenants can legally record conversations with landlords — but getting usable evidence requires more than just hitting record. This episode breaks down four specific conversational engineering techniques that turn mumbled complaints and ambiguous statements into clear, admissible admissions. From the "Repeat After Me" maneuver to the "Clarifying Q...

Daniel downloaded a horn soundboard app and ended up asking a surprisingly deep question: can a normal adult learn to play the bugle wake-up call, and how long would it take? We explore whether militaries still use live buglers, why the Swiss Army prefers brass over speakers in alpine terrain, and what the research says about adult skill acquisition. The answer might be faster than you think — abo...

After ten years and three nightmare landlords in Israel, one renter asks: Is this just bad luck, or is "the Ugly Israeli" a real, researchable phenomenon? We trace the historical roots of this cultural archetype — from the Six-Day War's psychological shift to the 1980s housing privatization — and examine psychometric data showing measurable empathy deficits in the demographic that dominates Israel...

When you see a wall of capital letters online, your first instinct is probably "this person is angry." But the reality is far more interesting. This episode breaks down the three distinct groups behind all-caps typing: older adults whose brains are reallocating cognitive resources, conspiracy theorists signaling epistemic frustration, and power users engaging in playful norm violation. We explore...

When you're up against someone who holds all the cards, sometimes the smartest move is to look confused. This episode unpacks "strategic incompetence" — a social engineering technique where you deliberately lower someone's guard by letting them underestimate you. We explore the three psychological mechanisms that make it work (overconfidence effect, halo of power, and cognitive load shifting), the...

What happens when the safest thing you can be online is a machine? This episode explores the strange reverse Turing test: humans deliberately adopting bot-like language patterns. From operational security to professional necessity, we break down the specific linguistic tricks — hedging phrases, symmetrical sentences, zero contractions — that make writing read as algorithmic. We also unpack the Mol...

Ever seen a website where you can browse passports like ETFs? CitizenX is a venture-backed platform that lets high net worth individuals purchase citizenship-by-investment (CBI) from countries like Dominica, Malta, and St. Kitts — with prices listed in fiat and crypto. In this episode, we explore the $25 billion global CBI market: who's buying (Chinese tech entrepreneurs, Russian oligarchs, Middle...

A sixteen-year-old accidentally carries a pocket flamethrower through airport security. DHS undercover tests show screeners miss weapons 95% of the time. So why do we still queue up for body scanners and liquid baggies? This episode unpacks the uncomfortable truth about security theater — a system optimized not for catching threats, but for making passengers feel safe. We trace the incentives that...

What happens when the privacy-maximizing advice civilians rely on becomes a red flag for intelligence agencies? This episode explores the high-stakes craft of building digital legends — fabricated online identities with years of plausible history, cross-platform consistency, and behavioral fingerprints that read as human. We trace how professionals maintain sock puppet farms, age accounts to look...

What happens when someone gets paid to lie their way past security guards, clone access badges, and talk their way into server rooms — and getting arrested is just a budgeted expense? This episode unpacks the strange, precarious world of physical penetration testing. From TSA Red Team operators smuggling mock explosives through checkpoints 95% of the time, to testers spending hours in holding cell...

When a landlord displaces a family with a six-month-old rather than fix a leak, is that just a bad person — or is the system designed to produce that outcome? This episode examines Israeli rental law's structural power imbalances, the psychology of petty sovereignty, and whether the landlord-tenant relationship itself selects for dehumanization. Drawing on data from the Central Bureau of Statistic...

In an era where the average US renter moves every 18 months, avoiding neighbor relationships is a rational response to a volatile market. This episode breaks down a full social engineering playbook for residential privacy, drawing on professional pretexting techniques used by penetration testers. Learn the three pillars of becoming a "satisfyingly boring" presence: question obfuscation, physical a...

What do you do when work, family, or circumstance lands you in a city that feels wrong? This episode tackles that tension head-on, using Jerusalem as a case study for a universal problem. We explore real research on place attachment — from a 2025 study showing that a single third place reduces dissatisfaction by 40%, to Hebrew University's cognitive mapping research, to Tel Aviv's surprising findi...

When Daniel sat down for pizza in Jerusalem, two women in full black veils walked past. Across the street, posters declared a long-dead rabbi the Messiah. He and his wife both felt the city was becoming more religiously homogeneous — but is that feeling backed by data? In this episode, we explore two tracks: the demographic numbers from the Jerusalem Institute for Policy Research showing how the c...

Why does buying a used platform trolley feel like a worse deal than buying new? This episode unpacks the hidden economics of second-hand marketplaces — from the endowment effect and hassle tax to the lemons problem that drives good sellers away. We explore why platforms optimize for listings instead of successful transactions, and how this creates a sustainability paradox where eco-conscious consu...