Glossary of Negative PR & AI Terms

Negative public relations (negative PR) is the strategic dissemination of damaging information about a target — a company, brand, or individual — through media, social media, review platforms, or AI answer engines. Unlike ordinary bad press, which is reactive and factual, negative PR is often proactive, coordinated, and designed to cause lasting reputational harm. It includes smear campaigns, planted stories, bot amplification, and AI-engine poisoning.

Dark PR (also called black PR) is a covert form of negative public relations in which the source of the attack is hidden or disguised. Dark PR operatives use anonymous accounts, front organizations, paid placements, and coordinated amplification to make damaging narratives appear organic. The defining feature is concealment of the sponsor — the target never knows who is behind the attack.

A smear campaign is a coordinated effort to damage the reputation of a target by spreading false, misleading, or damaging information across media and social platforms. Smear campaigns typically combine planted stories, bot amplification, review bombing, and (increasingly) AI answer-engine poisoning. They differ from legitimate criticism in their coordination, bad faith, and use of fabricated or distorted information.

Astroturfing is the practice of masking the sponsors of a message to make it appear as though it originates from grassroots participants. In negative PR, astroturfing creates the illusion of organic public sentiment against a target — fake reviews, coordinated social media comments, and fabricated community opposition. It is designed to make a coordinated attack look like spontaneous public consensus.

A planted story is a negative news article placed in a credible publication by a third party — often a competitor, short-seller, or coordinated operative — and then amplified through social media and secondary coverage. The credibility of the host publication lends weight to the claims, making planted stories one of the most damaging forms of negative PR and among the hardest to retract.

An AI hallucination is a confident but false output generated by an AI model — presented as fact when it is fabricated. In the context of reputation, AI hallucinations occur when ChatGPT, Gemini, Perplexity, or Google AI Overviews generate false claims about a person, company, or brand. Because the output appears authoritative, hallucinations can cause reputation damage even when no human source ever made the claim.

AI answer-engine poisoning is the deliberate seeding of false or damaging information into the source corpus that AI models (ChatGPT, Gemini, Perplexity, Google AI Overviews) use to generate responses. Once absorbed, the false information is repeated confidently to anyone who asks about the target — silently, persistently, and at scale. It is one of the most insidious forms of modern negative PR.

A deepfake is synthetic media (video, audio, or image) generated or altered using AI to depict a person doing or saying something they did not. In negative PR, deepfakes are used to create damaging 'evidence' — fabricated statements, actions, or appearances. They are especially dangerous because visual evidence is instinctively believed and shared faster than it can be debunked, even by sophisticated audiences.

A synthetic media attack is the deployment of AI-generated content — deepfakes, AI-written articles, fabricated screenshots, or cloned audio — as part of a coordinated reputation attack. Unlike traditional negative PR, synthetic media creates evidence that never existed. The attack is often timed to a sensitive moment (earnings, deal, election) for maximum damage before the fabrication can be exposed.

The source corpus is the collection of documents, articles, and data that an AI model uses to generate responses. When false or damaging information enters the source corpus (through publication, indexing, or deliberate poisoning), the AI will repeat it as fact. Source-corpus correction — identifying and addressing the originating and amplifying sources — is the primary method for fixing AI answer-engine damage.

Generative Engine Optimization (GEO) is the practice of optimizing content and structured data so that AI answer engines (ChatGPT, Gemini, Perplexity, Google AI Overviews) generate accurate, favorable responses about a brand or topic. GEO is the AI-era equivalent of SEO — it shapes how models describe you by influencing the source corpus and the structured signals they retrieve.

Disinformation is false information deliberately created and spread to cause harm. In negative PR, disinformation includes fabricated claims, forged documents, and AI-generated content designed to damage a target's reputation. It differs from misinformation (false information spread without harmful intent) in its deliberate, coordinated nature. Disinformation campaigns are increasingly amplified through AI answer engines.

Online Reputation Management (ORM) is the practice of monitoring, influencing, and managing the public perception of a brand or individual across search results, social media, review platforms, and AI answer engines. Modern ORM encompasses search-result management, review management, AI answer-engine optimization, crisis response, and reputation recovery. It is both a defensive (protecting against attacks) and constructive (building positive presence) discipline.

Reputation recovery is the sustained process of restoring a damaged reputation after a crisis or attack. Unlike acute crisis response (which contains the immediate damage), recovery addresses the lingering effects in search results and AI answer engines that can persist for months or years. Recovery requires authoritative content, structured-data signals, source-corpus correction, and consistent stakeholder communication.

A reputation shield (or reputation monitoring) is a continuous surveillance system that detects reputation threats — negative mentions, review patterns, social media sentiment shifts, and AI answer-engine changes — before they escalate into crises. Early detection is decisive in reputation defense: the difference between containment and amplification is often measured in hours.

Crisis communication is the strategic practice of managing public perception during and after a reputation crisis. It encompasses the holding statements, stakeholder briefings, media engagement, and AI answer-engine monitoring deployed when a crisis occurs. Effective crisis communication is early, factual, coordinated, and sustained — it addresses both the acute phase and the long recovery tail.

Defamation is a false statement of fact, published to a third party, that causes reputational harm. Libel is written defamation; slander is spoken. To be actionable, defamation generally requires falsity, publication, fault (negligence or actual malice for public figures), and demonstrable harm. Opinions are generally protected, even harsh ones. Defamation law varies significantly by jurisdiction.

Libel is defamation in a permanent or recorded form — written articles, social media posts, reviews, images, or published content. Because libel is recorded, it is generally easier to prove than slander and tends to cause more lasting damage. The permanence of written publication is what makes libel the most common actionable form of negative PR.

Slander is spoken defamation — false statements made verbally to a third party. Slander is generally harder to prove than libel because it is not recorded, and often requires proof of special damages (specific financial harm) unless the statement falls into a 'per se' category, such as accusing someone of a crime or professional incompetence.

A cease-and-desist letter is a formal demand that a party stop publishing defamatory or damaging content, typically sent before litigation. It is most effective when the claim is clear-cut and the recipient values accuracy. It can backfire if the claim is weak or the content is opinion, as recipients may publish it to generate 'they tried to silence us' coverage.

The right to be forgotten (RTBF) is a legal mechanism, primarily in the EU and UK under GDPR, that allows individuals to request search engines delist results that are outdated, irrelevant, or excessive. RTBF does not remove the source content — it makes it harder to find. It is limited in the United States, where First Amendment protections generally prevent forced delisting of accurate, public-interest information.

A SLAPP (Strategic Lawsuit Against Public Participation) is a defamation suit filed primarily to intimidate or silence a critic, rather than to remedy genuine reputational harm. Many jurisdictions have anti-SLAPP laws that allow early dismissal and fee-shifting. Ethical reputation defense avoids SLAPP-style tactics and focuses on genuinely defamatory content, not legitimate criticism.

Bot amplification is the use of automated or coordinated accounts to spread a message, making it appear to have organic public support. In negative PR, bot networks amplify damaging narratives until they cross the threshold of organic spread, at which point genuine users and media continue the propagation. Bot amplification makes a coordinated attack look like public consensus.

Review bombing is the coordinated posting of large volumes of negative reviews on platforms (Google, Yelp, Trustpilot, app stores) to damage a target's rating and visibility. Review bombing is often timed to a specific event and may combine genuine and fabricated reviews. Platforms have varying detection and response mechanisms for coordinated review attacks.

SEO poisoning is the deliberate manipulation of search engine results to make damaging content rank highly for a target's name or brand. Tactics include creating optimized negative content, building backlinks to it, and exploiting search engine ranking signals. SEO poisoning is particularly damaging because search results shape both human perception and AI answer-engine outputs.

Lawfare is the strategic use of legal action — lawsuits, regulatory complaints, or legal threats — as a weapon to intimidate, silence, or exhaust a target. In negative PR, lawfare is used to drain resources, create negative media coverage from filings, and force settlements. It is distinct from legitimate legal recourse, which seeks to remedy genuine harm.

Voice cloning is the use of generative AI to synthesize a convincing replica of a specific person's voice from a small audio sample, producing speech the person never said. In negative PR, cloned audio is used to fabricate damaging 'statements' from executives or public figures. Because audio carries an authenticity signal text lacks, cloned-voice attacks are quotable, embeddable, and persuasive — and they spread faster than they can be debunked.

Identity abuse is the use of a person's or organization's identity to create damaging content or gain unauthorized reach — fake accounts impersonating an executive, synthetic media depicting a real person, or account takeover to publish harmful posts from a legitimate handle. It is uniquely damaging because the content appears to come from the target, bypassing the audience's normal skepticism.

Content poisoning is the deliberate injection of false or misleading information into the sources that AI answer engines retrieve, index, and synthesize from. Unlike a single false claim, content poisoning targets the corpus the engine draws on, so the falsehood becomes part of every synthesized answer about the target. It exploits retrieval-augmented generation (RAG) and is one of the most insidious forms of modern negative PR.

Retrieval-Augmented Generation (RAG) is the method AI answer engines use to generate responses by retrieving current web content and synthesizing it into an answer, rather than relying solely on training data. Understanding RAG is central to AI reputation management: because the engine grounds its answer in retrieved sources, improving those sources — correcting inaccuracies, publishing authoritative content — changes what the engine synthesizes.

Google AI Overviews is Google's AI-generated summary that appears above organic search results for many queries. It synthesizes indexed content into a direct answer, often with citations. For reputation, AI Overviews matter because they appear at the top of the results page and are read as Google's answer — shifting the reputation surface from 'what ranks first' to 'what the summary says,' which is synthesized from the sources Google retrieves and weights.

Perplexity is an AI answer engine that synthesizes direct answers to user questions with prominent source citations, functioning as a real-time retrieval-augmented research tool. For reputation, Perplexity is significant because it updates quickly from fresh web content and makes its source set visible — which makes it one of the most tractable AI surfaces to monitor and correct through the cited sources.

ChatGPT Search is OpenAI's search feature within ChatGPT that retrieves and cites current web content to answer questions, complementing the model's training data with real-time retrieval. For reputation, ChatGPT Search matters because ChatGPT is one of the most widely used AI assistants, and its search feature shapes what millions of users read about entities. OpenAI recommends allowing its OAI-SearchBot crawler for visibility in ChatGPT search results.

Context collapse is the loss of original context when a fact, statement, or document is transported into a new setting, altering its meaning — usually making it more damaging. A name in a court filing is not a finding of fact, but once it appears, the document becomes a citation that downstream coverage and AI summaries treat as significant. The context that distinguished a mention from an accusation is lost at each step.

The Streisand effect is the phenomenon whereby an attempt to suppress, remove, or censor information draws wider attention to it, amplifying the very content the suppression was meant to hide. It is one of the most important concepts in reputation management because it describes the most common way well-intentioned responses backfire — a legal threat against a small article brings national coverage; a takedown demand causes republishing across platforms.

Permanence asymmetry is the principle that damaging content, once public, is effectively permanent, while corrections rarely travel as far as the original. The accusation reaches a large audience quickly; the correction reaches a smaller audience slowly and often never reaches those who saw the original. This asymmetry is the structural reason proactive reputation defense is more valuable than reactive repair.

Entity consistency is the state in which an entity's canonical facts — name, founders, ownership, headquarters, leadership — are identical and accurate across every source where they appear. AI answer engines reason about entities by reconciling attributes across sources; inconsistent facts produce low-confidence or contradictory summaries, and inconsistency favors whichever source is loudest — which is often not the entity itself.

Source authority is the credibility and weight a search or AI engine assigns to a given source when ranking results or synthesizing answers. In the AI-search era, source authority is the single most important variable in reputation: it determines which sources an AI engine cites and trusts, and therefore which version of an entity's story becomes the synthesized answer. Authority is earned, not declared, and cannot be faked at scale without detection.

Structured data is the machine-readable labeling of a page's facts using the schema.org vocabulary, typically in JSON-LD format, so search engines and AI answer engines can parse facts without inference. Organization, Person, Article, and FAQPage markup tell an engine exactly what each entity is. Without it, engines infer facts from unstructured text — and inference favors whichever source is most explicit, which is often not the entity itself.

A knowledge panel is the information box that appears for many entity searches, summarizing canonical facts sourced from Google's Knowledge Graph and structured data across the web. It matters for reputation because it is the first, authoritative summary a user sees for an entity, and it feeds the entity definitions that AI answer engines reconcile when they synthesize answers. An incomplete or incorrect panel leaves a vacuum that less authoritative sources fill.

Deindexing is the removal of a page from a search engine's index so it no longer appears in search results, distinct from source removal (which takes the page down entirely). Deindexing is the appropriate remedy when a page cannot be removed at the source but violates platform policy, privacy law, or a legal removal framework. It dramatically reduces reach even though the underlying page remains online.

Search result suppression is the strategy of publishing authoritative, relevant content so that damaging but lawful material ranks below it in search and AI results. Suppression does not remove the negative material; it reduces its visibility by ensuring accurate, authoritative content dominates the first page and the AI summary. It is the proportionate response to lawful, accurate, or opinion-protected material that cannot be removed.

Executive visibility is the managed prominence of a named leader — a CEO, founder, or public-facing executive — across search, news, social, professional profiles, and AI answer engines. It is the proactive counterpart to executive reputation defense: where defense protects against damage, visibility builds the authoritative presence that makes damage less likely and less severe. A leader with a strong visible presence is harder to damage because hostile narratives must compete with established authority.

Brand safety is the practice of ensuring a brand's advertising, content, and placements do not appear alongside contexts that could harm the brand by association — hate speech, misinformation, or controversial material. It matters for reputation because adjacency shapes perception: a brand repeatedly appearing alongside harmful content is perceived, however unfairly, as endorsing it. Association is automatic in automated ad systems, and the reputational risk is association.

A reputation audit is a structured assessment of how an entity is currently represented across search, news, social, reviews, AI answer engines, and public records. It is the starting point of any reputation program: it establishes the baseline, identifies threats and gaps, and prioritizes the work. The output is a prioritized action plan ranked by authority, reach, and risk — converting reputation management from a reactive activity into a measured, strategic program.