EP 04 · The Kobe Rule Shadow Metrics Media as Weapon

The Data Center

The barn has a server room. Dialog's chairman founded the two most important companies in the consumer location-data economy. Its attendees include the senators and regulators who have not regulated them. The behavioral futures market grades every body in real time — including yours.

YOUR BEHAVIOR IS THE RAW MATERIAL. YOUR FUTURE IS THE PRODUCT. IT IS SOLD WITHOUT YOUR KNOWLEDGE.

◆ The Arc So Far

Episode One: the livestock grammar. Episode Two: the public receipts. Episode Three: the sign factory that erases them. Episode Four enters the infrastructure beneath all of it — the data-broker ecosystem that grades, prices, and predicts every body in the system. This is where the Kobe Rule becomes automated. And this is where Dialog's founding becomes inseparable from the story.

Auren Hoffman is Dialog's chairman. He co-founded the network with Peter Thiel in 2006. He is also the founder of SafeGraph, a location-data brokerage that sold GPS movement traces of millions of Americans — including visits to abortion clinics, places of worship, and medical facilities — and LiveRamp, an identity-resolution company that matches anonymous data points back to named individuals across platforms.

At Dialog's 2026 retreat in Dublin, the leaked attendee directory placed Hoffman alongside Treasury Secretary Scott Bessent — who oversees financial data regulation — and Senator Ted Cruz, who chairs the Commerce, Science, and Transportation Committee responsible for FTC oversight of data privacy. The executives who sell behavioral data were in the same room as the officials who have so far declined to meaningfully regulate the sale of behavioral data. They used personal email addresses to avoid public-records law.

This is not an allegation of corruption. It is a description of access so structural it has become invisible.

◆ Subject File — Dialog Network

Dialog: What the Leak Confirmed
Source: Wired, June 2026. Internal documents inadvertently exposed online.
Founded
2006
By Peter Thiel (Palantir co-founder) and Auren Hoffman (SafeGraph, LiveRamp founder). No website. No public membership list. 20 years of operational secrecy.
2026 Attendees
222 registered
Trump administration officials, 2 U.S. senators, 6 PayPal Mafia members, former Middle East intelligence chief, sitting U.S. ambassador, surveillance and data-broker company founders — alongside their direct regulators.
Communication Method
Personal email
Government officials communicated via personal rather than official accounts — a documented strategy for avoiding FOIA and public-records requests.
Agenda Items
Cult-building, sex, WWIII prep, matchmaking
Per Wired's reporting on the leaked documents. The meeting also included substantive policy sessions on AI, health care, and geopolitics.
DC Expansion
Virginia campus purchased, 2025
Dialog purchased land near Langley and the Pentagon in 2025. Per Axios: "signals a longer-term desire to hold sway in Washington well beyond this administration."
Regulatory Overlap
Direct. Documented.
Palantir co-founder Joe Lonsdale listed alongside the House committee member who oversees Palantir's contracts. Hoffman listed alongside the FTC's direct oversight chain. Off the record.

The Behavioral Futures Market

In 2019, Harvard Business School professor Shoshana Zuboff named a new economic logic in her book The Age of Surveillance Capitalism. Her definition has not dated: "Surveillance capitalism unilaterally claims human experience as free raw material for translation into behavioral data. Although some of these data are applied to product or service improvement, the rest are declared as proprietary behavioral surplus, fed into advanced manufacturing processes known as machine intelligence, and fabricated into prediction products that anticipate what you will do now, soon, and later."

Those prediction products are then traded on behavioral futures markets — the real-time bidding ecosystem, the data-broker supply chain, the identity-resolution infrastructure. The market grades every body in the system continuously: your location, your attention, your emotional state, your likely political behavior, your health inferences, your creditworthiness. Dialog's chairman built two of the most important nodes in this grading infrastructure.

◆ The Vocabulary of the System (Zuboff, 2019)

Behavioral Surplus
The data exhaust produced by human activity — clicks, locations, searches, purchases, pauses, attention — beyond what is needed to deliver the service. This surplus is not disclosed. It is extracted without meaningful consent and fed into prediction engines.
→ SafeGraph's "Movement Panel" collected bidstream GPS data from millions of apps — data users never knowingly shared with SafeGraph.
Prediction Products
The outputs of machine-intelligence processing applied to behavioral surplus: scores, profiles, and forecasts about what individuals will do, buy, vote, believe, or feel. These products are sold to advertisers, employers, insurers, governments, and political campaigns.
→ Mobilewalla sold location data enabling targeting by political affiliation, religious attendance, and medical condition. FTC intervened in 2025.
Behavioral Futures Markets
The real-time bidding exchanges and data-broker networks where prediction products are traded. At each page load, an auction occurs in ~100ms: your inferred identity, location, emotional state, and behavioral score are broadcast to hundreds of bidders, a winning advertiser is selected, and the ad is served. You are not the customer. You are the inventory.
→ RTB market size: $21.4B in 2024, growing to $70B by 2035. The auction runs billions of times per day.
Big Other
Zuboff's name for the distributed surveillance architecture — not a single Big Brother state, but a network of private actors exercising unprecedented knowledge about and control over human behavior, operating largely outside democratic oversight. "A ubiquitous digital architecture: a 'Big Other' operating in the interests of surveillance capital."
→ Dialog's attendee list — data-broker executives alongside their direct regulators, off the record — is one node of the Big Other's governance layer.

"Google is to surveillance capitalism what General Motors was to managerial capitalism. The institutionalizing practices of Google are the primary lens — they produce a distributed and largely uncontested new expression of power: Big Other."

— Shoshana Zuboff, "Big Other: Surveillance Capitalism and the Prospects of an Information Civilization" (SSRN, 2015)

How the Grading Happens: Real-Time Bidding

Every page load triggers an auction that broadcasts your behavioral profile to hundreds of bidders. The entire process completes in under 100 milliseconds. Click each step to read what actually happens.

The 100ms Auction
What happens between you loading a page and seeing an ad
1
You load a page 0ms
Your browser sends a request to a publisher (a news site, an app, a blog). Alongside the content request, a signal is sent to the publisher's Supply-Side Platform (SSP) — the software that manages their ad inventory.
2
Your profile is assembled ~5ms
The SSP queries data brokers — including identity-resolution services like LiveRamp — to enrich your device ID with a behavioral profile: inferred age, income, health interests, political lean, recent location history, purchase intent. This happens without your knowledge or meaningful consent. The profile may contain hundreds of data points sourced from dozens of brokers.
3
Bid request broadcast ~10ms
The ad exchange broadcasts a bid request to hundreds of Demand-Side Platforms (DSPs) — the software used by advertisers to buy impressions. The bid request contains your enriched profile. Per EFF research, SafeGraph collected this "bidstream" data — the profile broadcasts from failed bids — to build its own behavioral database. Data is extracted even when no ad is served.
4
Auction and winner ~80ms
Advertisers submit bids. The winning bid — the highest price for your predicted attention and behavior — is selected. The FTC found that Mobilewalla retained data from failed bids, in violation of RTB exchange terms. The data extracted from losing bids is economically valuable: it maps which types of people appear at which publishers, at what times, in what locations.
5
Ad served and profile updated ~100ms
The winning ad loads. Your response — whether you click, how long you hover, whether you convert — is fed back into your behavioral profile, refining the prediction model. The cycle runs again on the next page load. Across the web, this auction runs billions of times per day. The global RTB market was $21.4B in 2024 and is projected to reach $70B by 2035.

The Scale of the Grading Infrastructure

$303B
Data broker market, 2024
Projected to reach $332B in 2025 and $479B by 2029, growing at ~9.8% CAGR. North America is the largest market.
Yahoo Finance / Data Broker Industry Analysis, 2025
$512B
Projected market size, 2033
Grand View Research projects the data broker market at $512.45B by 2033 — nearly doubling in under a decade, driven by AI-enabled behavioral modeling.
Grand View Research, 2024
$21.4B
Real-time bidding market, 2024
The auction layer where behavioral profiles are traded per impression. Projected to reach $70B by 2035 at 11.37% CAGR. Runs billions of auctions per day.
Market Research Future, 2025
0
Informed consent transactions
Bidstream data collection — harvesting behavioral profiles from failed ad bids — occurs without users knowing their data was broadcast in the first place. Consent is structurally impossible to give.
EFF / FTC enforcement actions, 2022–25

The Convergence: Dialog × Data Brokerage

The structural connection between Dialog's founding and the behavioral data economy is not coincidental. It is the same network, serving the same function: grading and predicting people, without accountability.

Two Infrastructures, One Grammar
Dialog (private elite coordination) and the data-broker ecosystem (behavioral surveillance) run on the same operating logic: collect without disclosure, grade without consent, coordinate without accountability.
Dialog Network
Auren Hoffman — Chairman. Founder of SafeGraph (location data) and LiveRamp (identity resolution).
Peter Thiel — Co-founder. Palantir co-founder. Government surveillance infrastructure.
Joe Lonsdale — Palantir co-founder. ICE case management. Pentagon data fusion.
Scott Bessent — Treasury Secretary. Oversees financial data regulation. Also a 2026 attendee.
Ted Cruz — Chairs committee overseeing FTC data privacy enforcement. Also a 2026 attendee.
Data-Broker Ecosystem
SafeGraph — Sold GPS movement data including abortion clinic visits. Used bidstream extraction. Agreed to stop selling Planned Parenthood visitor data after EFF reporting.
LiveRamp — Identity resolution: connects anonymous device IDs to named individuals across platforms. Core infrastructure of the behavioral futures market.
Mobilewalla — FTC action, 2025: collected location data enabling targeting by political affiliation, religious attendance, medical condition. Used failed-bid data in violation of RTB terms.
FTC — Brought actions against Gravy Analytics/Venntel and Mobilewalla in Dec 2024. Cruz's committee has formal oversight of FTC policy and budget.
◆ The Structural Point

Dialog's chairman built the identity-resolution infrastructure (LiveRamp) used to turn anonymous behavioral data into named profiles. Dialog's 2026 retreat placed him alongside the senator who chairs the committee overseeing the FTC — the agency that has been attempting to regulate exactly this practice. The meeting was off the record. Personal email only. No public disclosure required.

Research Annotation

The SafeGraph / abortion clinic data case is worth dwelling on because it makes the Kobe Rule's livestock grammar visible in its most precise form. Location data revealing visits to abortion clinics, churches, mosques, and medical facilities is not abstract. It is a behavioral grade attached to a body — a record of where a specific person went, when, and how often. That grade is then sold to anyone willing to pay: political campaigns, insurance underwriters, employers, law enforcement.

SafeGraph's defense — "we only sell data about physical places, not individuals" — is the exact linguistic operation the Sign Factory (Episode Three) described: renaming the product to erase the supply chain. The data is about places in the same way that a cattle tracking system is about locations: technically true, functionally a surveillance record of individual bodies.

The FTC's enforcement actions against Gravy Analytics/Venntel and Mobilewalla in December 2024 represented the first time the agency directly applied Section 5 Unfairness to the sale of sensitive location data. The orders were meaningful — prohibiting sale of sensitive location data, requiring deletion of historical data. But a federal court had previously dismissed an FTC action against Kochava, ruling that the FTC had not adequately alleged concrete consumer injury from location data exposure.

The enforcement gap is structural: the legal standard for "substantial injury" under Section 5 requires specific, demonstrated harm — not the systemic extraction of behavioral surplus without consent. The data-broker industry continues to operate at $300B+ scale in the space between the FTC's authority and the injury standard the courts require it to meet. Dialog's 2026 attendee list placed the people building this infrastructure alongside the people who set the standards for the agency trying to regulate it.

You Are the Livestock. So Are They.

Here is where the Kobe Rule closes the loop on its own metaphor. The behavioral futures market does not make an exception for the premium pen. The data-broker ecosystem grades Davos attendees' location data just as readily as anyone else's. The RTB auction does not check your net worth before broadcasting your profile to hundreds of bidders. The infrastructure is indifferent to the human hierarchy it serves.

This is Varoufakis's technofeudalism point, made concrete: the cloud capital — the behavioral data infrastructure, the prediction markets, the algorithmic grading systems — is not owned by the Dialog attendees. It is owned by the platforms and exchanges and data architectures that the Dialog attendees themselves depend on. The Wagyu steer has a TraceBeef ID. So does everyone in the room.

What the premium pen gets is influence over the grading criteria — not immunity from grading. Dialog's chairman shapes what counts as valuable behavioral data. Palantir's founders shape which behavioral data the government buys. The senators in the room shape which regulatory definitions protect or expose the system. They are not outside the livestock grammar. They are writing the breed standard.

"With little resistance from law or society, surveillance capitalism is on the verge of dominating the social order and shaping the digital future — if we let it."

— Shoshana Zuboff, The Age of Surveillance Capitalism (Harvard Business School, 2019)
Research Annotation

The most important thread to carry forward into the remaining episodes: Dialog is not primarily a social club. It is a governance structure operating outside formal accountability. The combination of data-broker executives, surveillance-infrastructure founders, intelligence officials, and the legislators who control the agencies regulating all of the above — meeting annually, off the record, in a building they now own near Langley — is a parallel policy process.

The formal democratic process (congressional hearings, FTC rule-making, legislative markup) is the public-facing layer. The informal coordination layer — Dialog, Davos side rooms, foundation-funded think-tank networks — is where the parameters of the formal process are set in advance. What gets regulated, at what pace, by what standard, with what enforcement budget: these are outputs of the informal layer, ratified by the formal one.

Episode Five follows one specific mechanism of this process: the revolving door at the regulatory agencies themselves — the career paths of the officials who move between writing the rules and working for the companies the rules govern. The data center runs because the doors keep turning.

The Grading Is Automated. Next: The Keepers of the Gate.

Episode 05 maps the regulatory revolving door in detail — the career paths of the officials who write the rules, then join the companies the rules govern. The barn door swings both ways.