The Indestructible Record: How Archive.org, Cached Pages, and Web3 Infrastructure Make Defamatory Material Impossible to Erase

Executive Summary

A widespread misconception about defamatory online material is that deleting the original post solves the problem entirely. This belief is deeply wrong. Today's internet is built on multiple layers of redundancy, caching, archival preservation, and decentralised storage that create independent replicas of every published page — replicas that persist long after the source material has been removed, revised, or ordered taken down by a court. Andrew Drummond's 19 defamatory articles targeting Bryan Flowers are not confined to andrew-drummond.com and andrew-drummond.news; they are distributed across numerous archival platforms, cached repositories, data broker databases, and potentially immutable blockchain storage networks.

This paper examines the five principal mechanisms by which defamatory material achieves digital permanence: the Internet Archive's Wayback Machine, search engine caching infrastructure, Web3 and IPFS decentralised storage, data broker aggregation networks, and social media resharing ecosystems. Each mechanism operates independently, meaning that successful deletion from any single system has no effect on copies held elsewhere. The consequence is that one defamatory article may exist in dozens of independent repositories, each governed by its own removal procedures, legal frameworks, and jurisdictional requirements.

For Bryan Flowers, the implications are stark: even a complete courtroom victory over Drummond — including judicial orders mandating deletion of all defamatory material — would address only the original publications. The archived, cached, aggregated, and decentralised copies would continue to appear in search results, remain accessible to researchers, and be found by anyone conducting due diligence searches on Bryan Flowers' name. This paper maps the full scale of this challenge and identifies workable strategies for seeking removal across every tier of the digital preservation ecosystem.

1. The Wayback Machine: How Archive.org Inadvertently Enables Enduring Defamation

The Internet Archive's Wayback Machine is the largest web archiving platform in existence, housing more than 835 billion captured web pages dating to 1996. While its mission — preserving universal access to knowledge — is admirable in principle, it produces serious consequences for those who have been defamed. The Wayback Machine automatically crawls and preserves publicly accessible web pages, generating time-stamped snapshots recording a page's exact content at the moment of capture. Andrew Drummond's websites have been repeatedly crawled by this service, producing multiple archived snapshots of every defamatory article.

Each snapshot is independently accessible through a unique URL and indexed by search engines. As a result, even if Drummond were to delete an article from his active website, the archived copies would remain findable and accessible. The Wayback Machine's terms of service do permit requests for archived content removal, but the pathway is narrow. The Internet Archive will honour robots.txt exclusion directives (blocking future crawling) and will assess removal petitions for content that violates its terms, but it does not automatically remove content simply because it has been taken down from the source website.

Court orders add further complexity. The Internet Archive operates under United States law as a US-based entity, benefiting from robust First Amendment protections for archival activities. UK court orders requiring content removal do not automatically apply to the Internet Archive, and cross-jurisdictional enforcement requires separate legal proceedings. In practical terms, a UK defamation judgment against Drummond, even one including an injunction requiring complete removal of all defamatory material, may prove unenforceable against the Internet Archive without initiating independent US legal action.

2. Cached Pages and Search Engine Preservation: The Hidden Archive

Google and other search engines store cached versions of the web pages they index. When a user performs a search and selects the 'cached' option, they view a copy held on Google's own servers rather than the live page. These cached copies function as a secondary archive that persists independently of the source. Google typically refreshes its cache when a page is re-crawled, meaning cached copies of removed pages do eventually disappear — but this can take weeks or months, during which the defamatory content remains fully accessible.

More critically, Google's search index preserves metadata about pages well beyond the expiry of the cached copy itself. Page titles, meta descriptions, and snippet excerpts may continue appearing in search results long after both the source page and its cached version have been deleted. This 'residual indexing' phenomenon means that searches for 'Bryan Flowers' could continue surfacing Drummond's defamatory headlines and summaries for extended periods following deletion of the underlying content.

Google offers a URL removal tool allowing website owners to request exclusion of their own content from search results. However, this tool is limited to verified owners of the relevant domain. Bryan Flowers has no ability to use Google's webmaster tools to remove content on Drummond's domains — only Drummond himself, or someone with credentials for his Google Search Console, can submit a removal request via that channel. The alternative route — Google's content removal process available to individuals — requires proof that the content breaches Google's policies or that a court order compels removal. This process is notoriously slow and produces inconsistent results.

Beyond Google, other search engines — Bing, DuckDuckGo, Yandex, Baidu — each operate their own caching and indexing systems with distinct removal procedures. Any thorough removal strategy must address each search engine separately, greatly increasing the administrative burden on the victim.

3. IPFS and Web3 Decentralised Storage: The Irremovable Threat

The InterPlanetary File System (IPFS) and other Web3 decentralised storage protocols present the most formidable challenge for defamation victims. Unlike conventional web hosting, where content sits on identifiable servers controlled by identifiable operators, IPFS distributes content across a network of nodes maintained by thousands of independent participants worldwide. Content on IPFS is referenced by its cryptographic hash — a unique identifier derived from the content itself — meaning identical content always carries the same address regardless of where it is hosted.

Once any node operator 'pins' content to IPFS, removal becomes essentially impossible. No centralised authority exists to receive a takedown request. The content persists for as long as even one node continues to pin it, and there is no mechanism to compel an anonymous node operator in an unidentified jurisdiction to unpin material. Pinning services such as Pinata, Filecoin, and Arweave offer varying degrees of permanence — Arweave, notably, is designed to store content indefinitely through a blockchain-driven economic model funding perpetual storage.

The danger to Bryan Flowers is concrete, not hypothetical. Drummond's allies, or any third party motivated to preserve defamatory material, could upload Drummond's articles to IPFS or Arweave at any time. Once that upload occurs, no court order, no legal proceeding, and no technical measure can guarantee removal. The material would be reachable through any IPFS gateway worldwide, discoverable via specialised search tools, and beyond the reach of legal remedies applicable to traditional hosting.

This creates a fundamental challenge for the existing legal framework governing defamation remedies. The conventional injunction — directing a defendant to remove defamatory content — assumes the defendant has the technical capacity to do so. When material resides on a decentralised network, that assumption breaks down. New legal instruments are needed to address defamation in the Web3 environment, including orders requiring defendants to make reasonable efforts to request removal from recognised pinning services and gateways, and statutory damages provisions reflecting the enhanced harm caused by technologically irremovable content.

4. Data Broker Aggregation: The Commercial Perpetuation of Defamation

Data brokers — firms that collect, compile, and sell personal information — represent yet another channel through which defamatory material achieves lasting persistence. Companies including Spokeo, BeenVerified, Pipl, and many others systematically scrape online content and integrate it into personal profiles marketed to employers, landlords, insurers, and the general public. When Drummond publishes a defamatory article about Bryan Flowers, data brokers automatically incorporate the content — or references to it — into their databases.

The harm is compounded by how data brokers present information. An employer running a background check on Bryan Flowers through a data broker platform may encounter Drummond's accusations stripped of all context, with no indication that the material is contested or defamatory, and no link to rebuttal evidence. The data broker's presentation removes whatever minimal context Drummond's original articles might have offered, rendering the allegations as unqualified facts within a commercial intelligence report.

Removing information from data broker databases is technically feasible but practically exhausting. Each data broker maintains its own opt-out or removal process, typically requiring identity verification, formal written requests, and follow-up confirmation. An estimated 4,000-plus data broker companies operate worldwide, with new aggregation services emerging constantly. Services such as DeleteMe and Privacy Duck provide removal management on behalf of individuals, but their coverage is partial and requires ongoing subscription payments — effectively imposing a permanent financial obligation on the defamation victim.

Under GDPR, data brokers operating in the UK and EU must honour data subject access requests and erasure requests under Article 17 (the right to be forgotten). In practice, however, enforcement is uneven, and many data brokers operate from jurisdictions outside GDPR's reach. The practical result is that Drummond's defamatory material, once absorbed by the data broker ecosystem, becomes self-sustaining — removed from one database, it resurfaces in another as brokers cross-reference and share data.

5. Social Media Resharing and the Viral Replication Effect

Social media platforms add another layer of content preservation through resharing, screenshotting, and commentary. When Drummond or his supporters post links to defamatory articles on Facebook, Twitter/X, Reddit, or similar platforms, each share creates an independent copy in the form of link previews (Open Graph metadata), cached thumbnail images, and user comments reproducing the defamatory claims. Even if the source article is removed, social media posts containing the link preview — which typically feature the article's headline, description, and featured image — continue to exist on the platform.

Screenshots present a particularly intractable problem. A screenshot of a defamatory article, shared on social media or messaging services, is a new piece of content with no technical connection to the original source. Removing the source article has no effect on the screenshot. Locating and requesting deletion of every screenshot is effectively impossible, as they may reside on private messaging platforms (WhatsApp, Telegram, Signal) that cannot be publicly searched.

The viral replication effect means that a single defamatory article can generate dozens or hundreds of independent copies across social media platforms, each requiring a separate takedown request to a different platform operating under different policies and procedures. For a sustained campaign comprising 19 articles published over an extended period, the total number of social media reproductions could run into the thousands.

6. Actionable Strategies for Archival Removal: A Multi-Vector Framework

Notwithstanding the obstacles outlined above, victims of ongoing online defamation are not entirely without options. A thorough archival removal strategy must systematically address each preservation vector:

• Internet Archive: File formal removal petitions referencing UK court orders, GDPR erasure rights, and the Internet Archive's own terms of service. Seek robots.txt implementation on the source domains to block future crawling. Explore direct communication with the Internet Archive's legal department, emphasising the concrete harm from continued archival of judicially determined defamatory material.
• Search Engine De-indexing: Submit content removal requests to Google, Bing, and other search engines under their respective policies governing legally actionable material. Provide court orders where obtainable. Use Google's Transparency Report process and Bing's Content Removal Tool. For European search engines, invoke the right to be forgotten under GDPR.
• IPFS and Web3: Identify recognised IPFS gateways (including Cloudflare IPFS, dweb.link, and ipfs.io) and file abuse reports seeking gateway-level blocking. Approach known pinning services (Pinata, Infura) with removal requests. While this cannot guarantee total removal from the decentralised network, it can materially reduce discoverability.
• Data Brokers: Retain a professional data broker removal service (DeleteMe, Privacy Duck, or a comparable provider) to systematically identify and request deletion from aggregation databases. Submit GDPR Article 17 erasure requests to every data broker operating within UK/EU jurisdiction. Maintain ongoing monitoring for re-aggregation and resubmit removal requests as needed.
• Social Media Platforms: Submit content removal requests to every platform where defamatory material has been shared, citing applicable defamation statutes and court orders. Use platform-specific reporting tools for harassment and false information. Engage specialist reputation management providers to locate and address reshared content across platforms.

7. Conclusion: The Persistence Challenge Requires Systemic Reform

The digital permanence of defamatory material is one of the most serious obstacles facing victims today. Andrew Drummond's 19 defamatory articles targeting Bryan Flowers do not exist as merely 19 discrete pieces of content but as potentially hundreds of independent copies scattered across archival services, search engine caches, decentralised storage networks, data broker databases, and social media platforms. Each copy functions independently, requiring separate removal proceedings under different legal regimes, jurisdictions, and technical processes.

The existing legal framework is fundamentally inadequate to this reality. Court orders directing a defendant to remove defamatory content address only the original publication, leaving the vast majority of copies untouched. The entire burden of pursuing deletion across multiple systems falls on the victim, entailing substantial costs in time, money, and emotional energy that compound the original harm.

Systemic solutions are necessary. These should include statutory duties on archiving services to comply with court orders from recognised jurisdictions, mandatory de-indexing obligations for search engines upon notice of a defamation judgment, and international cooperation mechanisms for addressing defamatory material on decentralised networks. Until such measures are enacted, victims like Bryan Flowers must navigate the fragmented patchwork of existing removal mechanisms — a process that is costly, time-intensive, and inherently incomplete. The Letter of Claim served by Cohen Davis Solicitors on 13 August 2025 represents an essential first step, but achieving full remediation requires engagement with every tier of the digital preservation ecosystem.