Bill Thompson, a former military intelligence operator and hacker who built offensive cyber tools for special operations, explains how government surveillance actually works, why he refuses to use Apple products, how AI is compressing hacking timelines from years to hours, and how he applied military-grade target analysis to build Spartan Forge — a hunting app that gives civilians 5cm-resolution mapping, LIDAR ground penetration, and a specialized AI assistant.
Why Bill Avoids Apple and Prefers Android
Bill chooses Android because its open-source project (AOSP) lets him inspect the OS, develop exploits, and dump his own phone for forensic analysis — essential when traveling to hostile countries or facing customs seizures where his device leaves his possession.
Apple’s closed architecture prevents this transparency; he would rather verify security himself than trust a corporation, even if closed systems theoretically hide vulnerabilities from attackers.
The Patriot Act was cleverly written to target “telephonic infrastructure” (microwave links, base transceiver stations), future-proofing itself for any new communication technology.
FISA (Foreign Intelligence Surveillance Act) operates as a secret “kangaroo court” near Fort Meade that approves surveillance requests without public oversight; the government only needs to assert a target is a foreign agent.
Five Eyes alliances allow countries to bypass domestic legal constraints (e.g., UK spying on US persons for the US) and share intelligence under a common umbrella.
Minimization procedures theoretically protect inadvertently collected US persons’ data, but the process moves “conveniently slowly,” becoming another vector for rights usurpation.
Government Incompetence vs. Free Market Agility
Legislation cannot keep pace with technology; government lacks the technical talent to understand, let alone regulate, emerging systems — the free market always attracts better talent.
DARPA is the rare competent government entity, funding moonshots for spin-off technologies (e.g., carbon fiber from a failed Mars rocket program), but its relevance has diminished as billionaires like Elon Musk can now fund big ideas privately.
Most government agencies are “self-licking ice cream cones” driven by budget execution, not mission success; generals get fired for unspent funds, creating perverse incentives to spend wastefully.
The Treasury’s payment infrastructure is a “Frankenstein’s monster” of 16 intersecting legacy systems, not a coherent database.
How Hacking Actually Works: Human vs. Technical Exploitation
95% of the time, recruiting a human insider (bribery, leverage, ideology) is cheaper and easier than technical hacking — classic spycraft using placement, access, motivation, and susceptibility.
Technical hacking (zero-days, implants) is reserved for hardened targets like encrypted satellite networks requiring physical proximity, decryption, and supply-chain interdiction.
Penetration testing (“ethical hacking”) is a victimless ransom: companies pay hackers to break their own systems and report vulnerabilities.
Forensic exploitation blends both: verifying human assets by dumping their phones to confirm they followed surveillance-detection routes and weren’t followed to meetings.
Pegasus and the Evolution of Spyware
Pegasus (likely Israeli-made) is a persistent implant that initially required a phishing link but evolved to zero-click exploits; it hides in obscure phone partitions that evade standard forensic triage.
The 2024 pager attack on Hezbollah (exploding pagers) exemplified years of supply-chain interdiction, corporate espionage, and patience — a “stroke of genius” in asymmetric warfare.
Threats constantly pivot: adversaries regress to primitive tech (pagers, pressure-plate IEDs) when advanced systems are compromised, forcing a perpetual cat-and-mouse cycle militaries are bad at because they’re built for symmetric warfare.
AI Is Supercharging Offensive Cyber Operations
AI compresses target development from months/years to hours: feed a device’s firmware, hardware specs, and network diagrams to a model and it identifies buffer overflows, unpatched IP tables, or admin password reuse.
Organized crime already trains private, guardrail-free models on proprietary exploit codebases using their own GPUs; state actors do the same.
Common criminal targets: phone access, banking credentials, Bitcoin wallets, and betting platforms.
What Intelligence Operators Understand About Human Behavior
Source recruitment follows a rigorous cycle (spot, assess, develop, recruit, handle) but the top 1% of operators combine that structure with “artistry” — rapid theory-of-mind, improvisation, and emotional control under chaos.
Cat-1 sources (e.g., a defense secretary) require artisans who can deduce motivations in 15 minutes and navigate counterintelligence traps; such operators are rare, usually 40+ years old.
High-value targets (Elon Musk, Trump family) are too visible to recruit directly; operators instead recruit adjacent people or build deep-cover companies as “backstopped” cover for long-term access.
Epstein-style operations are likely not state-run because professional agencies avoid flashy, compromised figures; the “glow” of visibility makes someone a poor asset.
The Self-Licking Ice Cream Cone: Government Incentive Structures
Agencies create threats to justify budgets (e.g., FBI infiltrating KKK meetings, white-supremacy divisions needing “wins” to survive); once the real threat fades, they manufacture it to keep funding.
This dynamic mirrors Big Pharma: perverse incentives keep people sick/dependent (statins over diet, Ozempic within the system) rather than solving root causes.
The result is public confusion and apathy — flooding the information space with competing narratives may be a feature, not a bug, paralyzing conviction.
Bill’s Path: From North Dakota Trailer Park to Military Intelligence to Spartan Forge
Bill joined the military to escape generational poverty, addiction, and a dead-end life in North Dakota; he re-enlisted after reading the Founders and committing to the American experiment.
His role was “the girthy part of the staff” — variable analysis for commanders, building tools (like Q for Bond) so tip-of-the-spear operators could kill bad guys.
Post-military, anxiety and depression drove him to build Spartan Forge: applying military target-analysis neural networks to deer movement prediction for hunters.
The app now offers 5cm-resolution aerial imagery (100x Google Maps), LIDAR ground penetration, historical imagery time-travel, offline maps, Blue Force Tracker for group hunts, and Cyber Scout — a hunting-specialized AI trained only on best-in-class sources.
Real-world impact: first responders used it during Hurricane Helene to find logging trails to reach cut-off homes; fathers and sons use it to create initiation-rite memories on public land.
Bill personally answers all social media messages (3,500+ after Rogan) as product owner, viewing direct user interaction as his most important role.
Where to Find Bill and Spartan Forge
Website: SpartanForge.ai; iOS/Android app stores; Instagram — Bill responds personally to questions, complaints, and hunting stories.
