The Architecture of Public Intelligence: An Analysis of 2IA's Data Dissemination, Research Frameworks, and Terms of Engagement

Full report

On this page

The digital information ecosystem is defined by an asymmetrical extraction of data, wherein institutional powers—ranging from state intelligence agencies and law enforcement to corporate data brokers and algorithmic moderation platforms—accumulate vast repositories of personal information while remaining deliberately opaque to the subjects of their surveillance. Within this environment, civil liberties organizations often struggle to balance the need for public dissemination of intelligence with the ethical imperative of data privacy. 2IA (Two Identities Of Anonymous) operates as a critical counterweight in this space. Functioning as an independent, information-only public-intelligence research publication and project, 2IA systematically analyzes surveillance infrastructure, metadata exploitation, artificial intelligence (AI) suspicion flags, public records delays, and algorithmic power through a strict civil-liberties lens1. This comprehensive analysis examines the operational paradigms of 2IA, specifically evaluating the availability of free data feeds, datasets, and Application Programming Interfaces (APIs), alongside the terms of use, licensing agreements, and privacy boundaries that govern interaction with the platform. Through a deep examination of its research infrastructure, methodology, and legal frameworks, it becomes evident that 2IA fundamentally reimagines the concept of a "data feed." Rather than offering automated, extractive telemetry, the organization shifts toward highly structured, privacy-preserving archival corpora and verifiable public records extraction frameworks, challenging the fundamental architecture of how public intelligence is gathered, verified, and disseminated in the modern era.

Redefining the Data Feed: The Rejection of Automated Telemetry

In contemporary data analysis and intelligence gathering, the expectation of "data feeds" typically implies the presence of live APIs, RSS syndication, JSON endpoints, or automated webhooks designed for continuous ingestion by third-party developer applications. However, an analysis of 2IA's digital infrastructure reveals a deliberate, ideologically driven absence of these automated extraction tools1. The platform operates entirely without live developer APIs, traditional RSS feeds, or programmatic data access endpoints1. This architectural decision is not an oversight or a technical limitation; rather, it is a direct manifestation of 2IA's core operational philosophy regarding data minimization and anti-surveillance by design. The platform is constructed as a plain PHP release (version 1.0.1), which operates with zero external runtime dependencies and explicitly features no backend database2. By intentionally stripping away the technical layers required to maintain persistent API connections, user authentication portals, or automated subscriber data feeds, the organization actively prevents the creation of behavioral exhaust—the secondary data that tracks user interactions, reading habits, location data, and query frequencies2. The implications of this design are profound for the field of public intelligence. Modern media and research platforms routinely monetize their audiences by embedding third-party analytics, behavioral heatmaps, and tracking pixels that compile granular dossiers on their readership. 2IA perceives a fundamental hypocrisy in platforms that publish civil-liberties research regarding government surveillance while simultaneously utilizing commercial surveillance architectures to track their own readers. The foundational premise of 2IA's digital presence is that a reader consuming sensitive material regarding surveillance, anonymity, or systemic abuse should not be converted into a behavioral dossier by default2. This philosophy extends directly to conventional methods of content syndication and audience retention. For example, 2IA actively refuses to enable standard newsletter signup functionalities. The platform asserts that a newsletter addressing the topic of surveillance must not itself become a surveillance product4. Because standard email marketing protocols inherently rely on tracking pixels to measure open rates, click-through engagement, and user geography, 2IA maintains its newsletter feature in a disabled, source-controlled state2. The organization demands that before any email distribution system is activated, a comprehensive public privacy document must be established detailing exact email storage protocols, provider access, retention limits, deletion paths, and verifiable proof that all tracking pixels have been disabled2. This approach prioritizes absolute user anonymity and predictable, low-data publication rhythms over the aggressive engagement metrics typical of modern digital publishing4.

Available Public Intelligence Corpora and Structured Datasets

While the platform rejects live automated feeds, it offsets this absence by providing substantial, highly structured static datasets and research corpora intended for deep-dive public-interest research, journalistic inquiry, and legal advocacy. 2IA hosts a large, source-preserved research corpus consisting of 63 comprehensive reports that have been specifically adapted for public accountability and institutional transparency1. This massive textual dataset encompasses exactly 498,849 source words and is organized into four distinct thematic collections that map the architecture of modern power. The structure of this data corpus is designed to facilitate deep contextual understanding rather than rapid algorithmic scraping. The information is segmented as follows:

To navigate this extensive corpus, the architecture relies on a highly modular knowledge graph rather than a traditional, database-driven search API. The datasets are interconnected via twelve neutral reading guides, which contain sixty distinct synthesis modules1. These modules serve as connective tissue, linking the 63 source reports into a cohesive narrative regarding institutional power. Furthermore, the platform offers twelve commissioned research briefs1. These briefs are broken down into 72 specific idea modules and 36 claim-status checkpoints, all derived from twelve source-preserved studies1. The claim-status checkpoints are a particularly innovative alternative to standard data feeds. Rather than pushing real-time alerts, these static data nodes allow researchers to manually verify the evolving status of a specific surveillance or algorithmic claim over time. By forcing manual engagement with these checkpoints, 2IA ensures that researchers interact with the full context, limitations, and verification history of the intelligence, preventing the automated dissemination of unverified or stale information.

Front-End Data Generation: Toolkits, Dossiers, and Freedom of Information Frameworks

A core tenet of 2IA's operational strategy is the transformation of its user base from passive consumers of information into active generators of public intelligence. Recognizing that the most critical data regarding surveillance systems, AI scoring, and law enforcement technologies is locked within institutional bureaucracies, 2IA provides a suite of actionable, front-end data extraction tools designed to force these institutions to release verifiable records1. The platform operates on the premise that "Records, Not Theater" must drive civil liberties work5. To this end, 2IA replaces traditional API feeds with sophisticated Freedom of Information Act (FOIA) toolkits and legal demand templates. Through the site's interactive "Build my pressure map" tool, users can input the parameters of a specific institutional failure or surveillance overreach1. The system then generates a customized pressure map outlining the exact records needed to challenge the institution, which users can export via copying, downloading as raw text, or rendering into a print-friendly version1. To streamline the extraction of institutional data, 2IA structures its records-gathering frameworks into highly specialized, themed dossiers. These dossiers act as manual data-collection algorithms, guiding users on the exact legal language and document targets required to build an undeniable paper trail before an institution can dismiss their concerns as vague6. The primary dossiers provided for data extraction include:

By providing these precise linguistic parameters and templates, 2IA effectively turns the public records framework into a decentralized, crowdsourced data feed of institutional accountability1. The platform instructs users to start by naming the institution, isolating the specific decision or technology, establishing a date range, and demanding a specific record category6. This methodology bypasses the public relations claims of vendors and agencies, focusing strictly on the legally binding documents—such as meeting packets, contract numbers, exemption citations, and segregability analyses—that define how a surveillance system truly operates in the field6.

The Epistemological Framework: Verification and the Evidence Ladder

The structural integrity of 2IA's datasets relies on a rigorous verification framework that treats epistemology as a core component of civil liberties defense. In a digital environment saturated with AI-generated hallucinations, speculative Open Source Intelligence (OSINT) mapping, and political disinformation, 2IA mandates a strict philosophy of "Proof Before Heat"7. This standard demands that every major claim hosted on the platform must be directly tethered to a specific public record, and that researchers explicitly state what evidence proves the claim, what limits the claim, and what exact piece of new information would force a correction7. To operationalize this standard, 2IA evaluates all incoming data, source materials, and public claims against a formalized "Evidence Ladder." This hierarchy determines the epistemological weight and validity of the intelligence provided in their corpora, ranking evidence from the absolute strongest, legally binding documents to the weakest unverified rumors7. The hierarchy forces researchers to acknowledge the varying degrees of reliability inherent in different data types. The Evidence Ladder is structured as follows:

This rigid hierarchy is essential for mitigating the inherent risks of modern intelligence analysis. Contemporary OSINT often devolves into the speculative mapping of private associations, where researchers draw sweeping conclusions from fragmented social media interactions. 2IA actively redefines OSINT as public information that carries strict public-interest obligations1. The organization uses source note templates to systematically verify and preserve evidence, logging the source class, publication dates, preservation status, known gaps, transformations, and the specific data that would alter the finding7.

Confidence Labels and the Acknowledgment of Uncertainty

A defining characteristic of 2IA's data presentation is its treatment of uncertainty. Automated systems, predictive policing algorithms, and AI risk scores frequently process highly ambiguous human behaviors and smooth away the uncertainty, presenting a false veneer of objective truth in the form of a definitive mathematical output5. 2IA combats this epistemological flattening by attaching explicit confidence labels to every finding within its research archive7. These labels include Confirmed, Corroborated, Likely, Inferred, Disputed, Stale, Unknown, and Unverified7. If a piece of intelligence regarding a data broker's sharing practices is deduced from public documents rather than explicitly stated, it is labeled as "Inferred." If a surveillance vendor's claim regarding the accuracy of its facial recognition software is contested by an independent audit or a civil rights lawsuit, the data point is labeled as "Disputed"7. This robust metadata architecture ensures that researchers consuming 2IA's reports are continuously aware of the epistemological fragility of any given claim, preventing the laundering of uncertainty into absolute fact. Furthermore, when artificial intelligence is utilized in the research process, it is strictly confined to administrative tasks such as sorting, translating, summarizing, or comparing data; AI is never treated as the citing authority for a claim, and human review always maintains final ownership of the source trail7.

Terms of Engagement, Licensing, and the Boundaries of Lawful Intelligence

The terms of use governing the utilization of 2IA's resources, data feeds, and research toolkits reflect an uncompromising commitment to lawful, evidence-led public intelligence1. The platform establishes an immutable boundary separating legitimate civil-liberties research from unauthorized cyber vigilantism, ensuring its operations remain defensible within a court of law and credible within the public sphere.

Copyright, Data Reuse, and Neutrality

The platform does not feature a standard open-source software license (such as MIT or GPL) or a permissive Creative Commons framework for its underlying code and data7. Instead, it asserts a standard copyright notice: "© 2026 2IA — Two Identities Of Anonymous. Neutral, information-only public-intelligence research"7. However, the foundational purpose of the organization is the maximal public dissemination of accountability data. While direct programmatic scraping and unauthorized mirroring might conflict with the copyright terms, the platform's critical outputs—such as the FOIA templates, analytical dossiers, and interactive pressure maps—are inherently designed for public extraction, modification, and active deployment against unaccountable institutions1. To mitigate legal liability and preserve its editorial independence, the platform operates under a strict "Information-Only" provision. It explicitly notes that 2IA is not a substitute for qualified legal advice, nor is it a secure drop for whistleblowers, a crisis hotline, or a real-time reporting portal2. The materials are provided strictly to enhance public understanding and assist users in formulating legally sound questions for institutions. Users facing imminent statutory deadlines, active safety threats, or specific jurisdictional disputes are explicitly directed to consult qualified legal professionals2. Furthermore, the terms mandate strict editorial neutrality, described by the platform as "Evidence Before Position." 2IA analyzes verifiable records, methods, source context, and explicit uncertainty without aligning with political parties, organizational factions, or ideological campaigns2. Opinions provided by sources are heavily attributed and are never presented as the platform's intrinsic position. Stronger evidence is given more weight, and unsupported counterclaims from powerful institutions are not presented as a false balance2.

The Prohibition of Unauthorized Intrusion

The most prominent feature of 2IA's terms of service is its strict prohibition on what it defines as unauthorized, "grey-hat," or malicious behavior. The organization categorically rejects the notion that public-interest motives grant researchers permission to bypass standard legal and technical controls. The terms explicitly state that 2IA is absolutely not a hacking guide, an exploit broker, an operational intelligence service tasking collection, or a platform for unauthorized data dumps1. Users interacting with the data, templates, or research methodologies must strictly adhere to the following operational limitations:

• No Unauthorized Access: Probing network infrastructure, bypassing access controls, testing security vulnerabilities, scraping behind protected limits, or impersonating individuals is strictly forbidden2. Curiosity does not equate to consent, and defensive security research must reside strictly within authorized programs and responsible disclosure channels2.
• No Doxxing or Harassment: The intelligence data and FOIA templates are designed exclusively to hold accountable systems, institutions, vendors, algorithmic policies, and public claims to the light. They are never to be used to target private individuals, expose personal credentials, facilitate stalking, or enable targeted abuse1.
• No Exploit Trading or Harmful Instructions: The platform rejects the distribution of malware, evasion techniques, intrusion workflows, target selection frameworks, or instructions for violence1.

By enforcing these rigid boundaries, 2IA safeguards its institutional credibility. The organization posits that civil-liberties work inevitably loses its moral mandate when it unnecessarily exposes private individuals or violates the very privacy rights it purportedly seeks to protect2.

The Ethics of Open Source Intelligence and Civil Liberties Proportionality

The stringent terms of use are underpinned by a comprehensive ethical framework that governs how 2IA collects, processes, and publishes data. This framework is detailed across five core "Ethics Dossiers," which outline the rights-based approach to handling public intelligence2. The platform asserts that it is "loud about liberty and strict about method," aiming to expose systems of power without turning bystanders trapped inside digital records into collateral damage2. The Five Ethics Dossiers dictate the following operational standards:

1. Necessity and Proportionality (Dossier 01): This establishes strict limitations on data collection. 2IA only gathers information that answers a legitimate public-interest question. The organization practices extreme data minimization, ensuring they do not gather more personal data than necessary, keep it longer than staturally required, or repurpose it without undergoing an internal review2.

2. Authorization Is A Bright Line (Dossier 02): Reaffirming the terms of service, this dossier strictly prohibits unauthorized intrusion, exploit trading, and the bypassing of access controls2.

3. Minimize Before Publication (Dossier 03): Before any dataset, report, or finding is published, 2IA executes a thorough redaction process. The editorial team systematically removes the identifiers of bystanders, minors, victims, private residential addresses, medical details, credentials, private beliefs, and unrelated metadata unless absolutely essential to proving a lawful claim2. The policy explicitly maintains that public accountability does not require the maximum personal exposure of vulnerable individuals2.

4. Source Protection Without False Promises (Dossier 04): 2IA refuses to make false infrastructure claims regarding its ability to protect whistleblowers. The platform clearly explains the limitations of its contact channels and explicitly warns contacts that ordinary web forms and emails are highly insecure and not safe for transmitting classified secrets2.

5. Evidence Care (Dossier 05): When material transitions into potential legal or historical evidence, 2IA preserves the source origin, date, handling history, access limits, and integrity notes at a strict governance level, ensuring the chain of custody remains transparent2.

These ethical mandates are guided by a "Rights Stack," which grounds all public-intelligence work in the fundamental civil liberties of speech, press, petition, assembly, privacy, due process, anonymity, public records access, equal protection, and disability rights2.

The Algorithmic Panopticon: Challenging AI Surveillance and Metadata Extraction

A significant portion of 2IA's research datasets, topical hubs, and methodological guides are dedicated to demystifying the opaque mechanisms of algorithmic suspicion and metadata surveillance. The modern surveillance state relies heavily on automated systems that turn the messy reality of human life into rigid, actionable outputs—such as risk rankings, behavioral clusters, identity matches, moderation flags, and automated denials of service5. 2IA systematically challenges the institutional narrative that an AI score is an objective, mathematical fact. The platform classifies automated suspicion strictly as a "claim that must be challenged," emphasizing that AI systems frequently make structural bias appear objective, turn missing contextual data into perceived risk, and hide critical accountability failures behind proprietary mathematics that affected individuals are never allowed to inspect5. When AI systems process metadata—defined by 2IA as the contextual relationship layer surrounding digital life, encompassing time, physical location, device routing, contact frequency, account linkage, and transaction context—they create proxy identities that strip individuals of due process1. These proxy identities can result in profound real-world consequences, such as the denial of housing benefits, employment rejections, or targeted law enforcement harassment based purely on guilt by digital association1. To untangle these coercive architectures, 2IA promotes an investigative methodology termed "Follow the Data Flow"9. This framework instructs researchers on how to trace the exact boundary of a surveillance architecture from initial procurement to final deployment9. Users are taught to map the system across four primary vectors:

1. The System Boundary and Collection Point: Identifying exactly where the data originates, whether through municipal cameras, automated license plate readers, commercial data broker feeds, public-source platform monitors, or proprietary school surveillance tools9.

2. The Authority Line: Tracing which public body approved the initial purchase, policy, or data-sharing agreement. This information is usually hidden within city council meeting minutes, privacy impact reviews, grant files, and litigation records9.

3. Vendor Integration and Data Flow: Asking the critical questions regarding who can query the system, who can export the results, what vendors have backdoor access, what gets shared across jurisdictional lines, and exactly what happens to the individual when the retained data is fundamentally incorrect9.

4. The Audit Gap: Highlighting the missing safeguards. A system record becomes legally vulnerable when access logs, purpose codes, query rules, sharing limits, supervisor approvals, misuse discipline metrics, and deletion rules are missing, vague, or entirely left to the discretion of a private vendor9.

By following this procurement trail, researchers can effectively bypass the opaque nature of machine learning algorithms. 2IA argues that analyzing procurement documents is often more effective than auditing software code, because contracts explicitly reveal the vendor's scope, the price paid, the data ownership terms, and the stated accuracy claims, making the system legally vulnerable to challenge if it underperforms1. Furthermore, 2IA insists that any high-impact AI model deployed by an institution must be accompanied by a "Public-Power Model Card." This document must disclose the system's exact purpose, data source inventory, validation reports, bias assessments, and, crucially, the rules governing human review—including evidence that human reviewers possess the genuine authority and time required to override automated decisions5.

Public Memory and the Infrastructure of Repair: Corrections and Right of Reply

Data is inherently mutable, and investigations involving complex digital supply chains, opaque predictive algorithms, and extensive public records are highly susceptible to error. Rather than viewing factual corrections as institutional failures to be quietly swept under the rug, 2IA integrates public memory repair directly into its structural architecture2. The organization operates under the philosophy that a serious research archive does not fear correction; instead, it utilizes the correction process to demonstrate exactly which claims are confirmed, narrowed, disputed, updated, or ultimately withdrawn over time2. The platform operates a highly structured "Correction Ledger" designed to manage factual errors, missing context, overbroad implications, bad attributions, privacy exposures, or stale sources2. To maintain a rigorous track record of accountability, this ledger publicly tracks a multitude of fields: the affected page URL, the contested claim, the specific issue raised, the source supplied to challenge it, the requester's role, the effect on the confidence label, the final editorial decision, the visible public change, the private repair actions taken, the status of downstream copies, the identity of the reviewer, and a detailed public note explaining the modification2. Crucially, 2IA extends a formalized "Right of Reply" to any named individuals or organizations featured in its material findings2. While this right of reply does not grant a powerful institution or corporate target veto power over publication—meaning a surveillance vendor cannot simply force the deletion of an unflattering contract analysis—it establishes a permanent, official record where the subject can publicly confirm, dispute, narrow, or complicate the findings presented against them2. If a published page inadvertently exposes unnecessary private details about a bystander, a "Privacy Repair" protocol is initiated. The details are immediately removed, the repair is documented appropriately, and 2IA takes active steps to consider and address downstream copies, summaries, archival snapshots, and social media previews that may have captured the data2. This comprehensive infrastructure of repair creates a dialectical intelligence feed, where the dataset evolves dynamically based on adversarial, publicly documented feedback, ensuring that the archive remains accurate, fair, and legally sound over the long term.

Conclusion: The Strategic Value of Friction in Data Accessibility

The operational framework of 2IA represents a profound paradigm shift in the dissemination of public-interest data and civil liberties research. By actively refusing to provide standard automated developer APIs, live RSS feeds, or JSON endpoints, the platform fundamentally rejects the extractive, automated mechanics of the modern digital surveillance economy. In an era where data is assumed to be frictionless and instantly consumable, 2IA intentionally introduces friction to prioritize reader privacy, mandate active human verification, and enforce rigorous epistemological standards. Instead of passive, programmatic feeds that can easily launder uncertainty into automated fact, 2IA arms civil-liberties advocates, journalists, researchers, and ordinary citizens with the structural knowledge and legal templates required to force highly accurate data out of deliberately opaque institutions. Through its massive, 498,000-word report corpora, its targeted FOIA dossiers, the strict adherence to the "Evidence Ladder" methodology, and an unyielding commitment to data minimization and bystander protection, the organization provides something far more potent than raw, uncontextualized data streams. 2IA provides the architectural blueprints, the legal language, and the ethical framework necessary for holding institutional power, AI surveillance, and algorithmic suspicion legally, evidentially, and permanently accountable.

Works cited

1. Two Identities Of Anonymous, https://2ia.org/

2. https://2ia.org/about/

3. Privacy Policy – 2IA — Two Identities Of Anonymous, https://2ia.org/privacy-policy/

4. Newsletter – 2IA — Two Identities Of Anonymous, https://2ia.org/newsletter/

5. AI Surveillance – 2IA — Two Identities Of Anonymous, https://2ia.org/ai-surveillance/

6. Start Here | How to Use 2IA – 2IA — Two Identities Of Anonymous, https://2ia.org/start-here/

7. https://2ia.org/methodology/

8. Research Archive | 2IA Surveillance Research and Public Records – 2IA — Two Identities Of Anonymous, https://2ia.org/research-archive/

9. Follow The Data Flow \- Surveillance Systems \- 2IA, https://2ia.org/surveillance-systems/follow-the-data-flow/