The document is real.
The quotation is accurate.
The photograph has not been altered.
And yet the way the material is released may still be designed to mislead, intimidate or cause harm.
This is the problem that malinformation was introduced to name.
It is also one of the least stable terms in the information-disorder vocabulary.
Definition
The influential 2017 Council of Europe report by Claire Wardle and Hossein Derakhshan distinguished three categories:
misinformation, disinformation and malinformation.
In that framework, malinformation involves genuine information shared to cause harm, often by moving information intended to remain private into the public sphere.
For this Knowledge Library, the term is used narrowly:
genuine or substantially accurate information strategically used, disclosed, reframed or decontextualized in a manner intended to cause harm or produce a misleading interpretation.
That definition should be read as a taxonomy aid, not as a settled scientific construct.
Why the category exists
Truth-value alone cannot describe every harmful information practice.
An authentic private address can be published to facilitate harassment.
A real photograph can be detached from its original date and used to imply something about a different event.
A genuine document can be selectively leaked so that one accurate fragment dominates interpretation while relevant context remains hidden.
A true personal fact can be disclosed with the explicit aim of intimidation.
These cases do not fit comfortably inside a simple “false information” category.
Malinformation gives analysts a word for the possibility that information can be epistemically genuine yet communicatively weaponized.
What malinformation is not
Malinformation is not any information that causes harm.
Accurate investigative journalism may damage a powerful person’s reputation because it reveals misconduct. Public-interest reporting should not become “malinformation” merely because someone dislikes its consequences.
It is not identical to doxxing, although some doxxing can fit the taxonomy.
It is not identical to privacy violation, because privacy can be breached without an informational strategy designed to mislead or harm a wider audience.
It is not whistleblowing by definition. Whistleblowing often involves genuine information disclosed because the public-interest case for disclosure outweighs confidentiality, even when institutions experience harm.
And it is not disinformation if the relevant content is genuine and the taxonomy in use reserves disinformation for false or misleading information.
These distinctions are ethically and legally consequential.
Harm is not self-defining
The word “harm” creates immediate difficulty.
Who decides what counts as harm?
A revelation can harm an individual while benefiting the public. Disclosure can violate privacy while exposing abuse. Publication can create security risk while also revealing wrongdoing.
This means that malinformation classification can become normatively loaded in a way that misinformation classification sometimes is not.
The analyst has to separate at least four questions:
Is the information genuine? Was the disclosure or framing intentional? What kind of harm was intended or reasonably foreseeable? Was there a competing public-interest justification?
Collapsing these questions invites misuse of the category.
Context can make true material misleading
A claim need not become false when context disappears.
Its implication can change.
A genuine statistic without its denominator can create the wrong impression. A real image shown with the wrong date can suggest an event happened recently. An accurate quotation can be clipped before the sentence that reverses its apparent meaning.
Some scholars would classify such cases as misleading information rather than malinformation. Others may treat them as disinformation when the contextual manipulation is deliberate.
This overlap is not a reason to pretend the terminology is settled.
It is a reason to record the classification rule.
Intent is difficult here too
The same evidentiary problem that affects disinformation affects malinformation.
Observers usually see the released material before they see the producer’s internal motives.
A private document posted online may have been leaked to expose wrongdoing, embarrass a rival, provoke harassment, support litigation or simply attract attention.
Intent should therefore be inferred cautiously and, where possible, from corroborating evidence such as planning records, coordinated distribution, explicit statements or repeated behavioural patterns.
The label should not be used merely because the outcome was harmful.
The empirical literature is comparatively thin
Misinformation research has large experimental, correction and intervention literatures.
Disinformation research spans communication, security, computational methods and political communication.
Malinformation has much less independent empirical development.
Many publications invoke the Wardle–Derakhshan taxonomy or use the term descriptively rather than testing a distinct malinformation mechanism.
A 2026 cross-sectional study by Pamirzad and Chen, for example, examines associations involving disinformation and malinformation exposure in an Afghan sample of 520 participants. The study is useful evidence that researchers are operationalizing the distinction, but its design and geography do not establish a general causal theory of malinformation.
This thinness is an important part of the page, not a flaw to hide.
Why the page still clears the independent-page threshold
The Institute allows the library to shrink when a planned entry cannot carry an independent research and editorial job.
Malinformation is close to that boundary.
It survives because it answers a question that misinformation and disinformation pages do not answer cleanly:
How should a discernment framework reason about informational harm when the underlying information is genuine?
It also provides a necessary bridge to privacy, context integrity, strategic disclosure, provenance and source ethics.
The page therefore earns its place as a qualified taxonomy reference, not as evidence of a mature standalone psychological construct.
Truth is not the only epistemic property that matters
An information system can preserve literal accuracy while damaging understanding.
Selective release can distort salience. Decontextualization can change implication. Privacy violations can alter power. Repetition can make a marginal fact dominate collective attention.
This suggests a broader principle for discernment:
Truth is necessary for many epistemic judgments, but not sufficient for evaluating the quality of an informational act.
Context, relevance, proportionality, provenance and intention can matter too.
That principle should not be used to weaken factual standards.
It should be used to understand cases in which factual accuracy coexists with manipulation.
Malinformation and journalism
The boundary with journalism needs special care.
Journalists routinely publish information that sources or institutions would prefer to remain private.
A taxonomy that labels every harmful truthful disclosure “malinformation” could become a tool for suppressing legitimate reporting.
Public-interest tests, professional ethics, proportionality, verification and legal protections therefore matter.
The Institute will not infer malinformation merely from unauthorized disclosure.
The question is not simply whether the information was private.
It is how the disclosure functions, what purpose can be evidenced, what context is preserved and what competing public interest exists.
Malinformation and personal privacy
At the individual level, the concept overlaps with harmful exposure of personal information.
A home address, intimate image, medical detail or identity attribute can be accurate and still be weaponized.
Here the epistemic analysis meets consent and safety.
Yet not every privacy violation belongs under malinformation. Legal and ethical frameworks may offer more precise categories.
The term should be used only when its information-disorder function adds explanatory value.
AI does not make truth harmless
Generative systems can synthesize and amplify truthful material too.
They can assemble scattered public data into invasive profiles, summarize private archives, identify patterns individuals did not expect others to infer, or reframe authentic material for targeted persuasion.
This raises questions adjacent to malinformation even when no fabricated content is introduced.
The relevant problem is not “AI misinformation.”
It may be information aggregation, privacy, contextual integrity or strategic disclosure.
Accurate naming matters because different problems need different safeguards.
Measurement and classification limits
There is no widely accepted malinformation scale.
Researchers may classify cases by truth status, privacy expectation, context shift, inferred intent and harm. Those dimensions are not standardized across fields.
Inter-rater agreement can depend heavily on the information available to coders.
A case may change classification when new evidence about provenance or intent emerges.
This means malinformation should generally be treated as an analytic classification with explicit criteria rather than a directly observed natural kind.
The independent-page decision in detail
The independent-page threshold matters most when a term is familiar enough to deserve explanation but not mature enough to support inflated authority.
Malinformation receives one point rather than two for independent source base because much of the literature inherits the 2017 taxonomy instead of building a large distinct empirical programme. It receives stronger scores for reader utility, graph value, Institute fit and unique editorial contribution.
The decision to KEEP is therefore conditional on the page remaining what it is now: a boundary-setting taxonomy reference with explicit limitations.
If future evidence review shows that researchers consistently collapse the term into broader information-disorder categories, the Institute should merge it rather than defend page count.
This is the kind of shrinkage rule that keeps the library honest.
Contextual integrity offers a neighbouring lens
Privacy scholarship provides another way to think about truthful information used harmfully: contextual integrity.
The core idea is that information flows are governed by contextual norms concerning actors, information types and transmission principles. A fact can be publicly knowable in one setting yet inappropriately transferred into another.
The Institute should not pretend that contextual integrity and malinformation are the same theory.
They illuminate overlapping territory from different traditions.
The comparison is useful because it prevents “harmful truth” from becoming purely subjective. Analysts can ask whether an information flow violates established contextual expectations, not only whether someone reports feeling harmed.
Public interest is not a simple exemption
Invoking public interest does not settle every disclosure dispute.
Journalists, regulators and courts may weigh proportionality, relevance, necessity, privacy and accountability differently.
The Institute is not a legal authority and should not use this page to resolve those disputes.
Its epistemic role is narrower: to insist that the purpose and context of truthful disclosure be analyzed separately from truth status.
That is already a meaningful improvement over a binary model in which “true” equals harmless and “false” equals harmful.
Decontextualization sits on a conceptual border
When an authentic image is shown with a false date, what exactly is true?
The pixels are genuine. The proposition implied by the presentation may be false.
Some analysts will classify the case as disinformation because the communicated claim is false. Others may emphasize malinformation because genuine content has been strategically redeployed.
This is not merely semantic fussiness.
It reveals that informational artifacts contain multiple levels: material authenticity, propositional meaning, contextual implication and communicative intention.
A robust taxonomy should state which level it is classifying.
Malinformation and aggregation
One of the most important emerging cases involves aggregation rather than disclosure of a single secret.
Many individual facts may be public. Combining them can create a profile, risk or inference that no one data point carried alone.
AI systems make aggregation cheap.
A person’s workplace, routines, family links, public photographs and professional history can be synthesized into something much more invasive than the original fragments.
Calling every such case malinformation would stretch the term too far.
But the phenomenon reinforces the page’s central insight: informational harm cannot be evaluated by truth value alone.
A conservative use policy
Because the term is contested, the Institute should use it conservatively.
A publication should label a case “malinformation” only when it states the taxonomy being used, establishes that the relevant information is genuine, explains the alleged harmful use, separates intent from outcome and notes any public-interest dispute.
Where those conditions cannot be met, more descriptive language is preferable.
For example: “authentic private information was published without consent” may be clearer than applying a disputed category.
Precision is especially important because labels can themselves become instruments of power.
Research priorities
A stronger malinformation evidence base would need comparative work across legal and cultural contexts, validated coding frameworks, inter-rater reliability studies, clearer distinctions from privacy violation and decontextualized disinformation, and research on how audiences interpret genuine information used strategically.
It would also benefit from work on AI-mediated aggregation and contextual collapse.
Until then, the Institute should resist the appearance of maturity that a polished definition can create.
The page can be useful precisely because it keeps the uncertainty visible.
The cost of a bad label
Misclassifying truthful disclosure as malinformation can carry real consequences.
It can delegitimize whistleblowers, discourage investigative reporting or allow institutions to portray embarrassment as informational harm. The reverse error matters too: treating strategically exposed private material as harmless merely because it is true can ignore intimidation and safety risks.
This asymmetry is one reason the term should trigger explanation rather than replace it.
Whenever possible, the Institute should describe the underlying act first: authentic private records were selectively released; a real image was recaptioned; public data were aggregated to identify an individual.
Only then should the taxonomy label be added.
Descriptive priority makes the page useful even if the terminology changes in future research.
What the evidence does not show
The evidence does not show that malinformation is a settled psychological construct.
It does not show that every truthful harmful disclosure belongs in the category.
It does not show that unauthorized disclosure is automatically malicious.
It does not show that harm alone reveals intent.
It does not show that existing malinformation findings generalize across cultures, legal systems and media environments.
And the taxonomy does not resolve conflicts between privacy, security, accountability, journalism and public interest.
What this changes for Inner Technology
Malinformation forces the discernment system to ask more than whether a statement is true.
The Institute therefore uses a four-part check:
Truth status. Context integrity. Provenance and intention. Public-interest and harm boundary.
This is Institute synthesis, not a validated malinformation instrument.
Its value is protective: it prevents a truth-versus-falsehood framework from becoming blind to strategic uses of genuine material.
The authentic document
The document remains real.
That is the beginning of the analysis, not the end.
A mature information ecology has to protect truth without pretending that every use of truth is benign.
It also has to protect journalism and accountability without allowing “harm” to become a convenient excuse for secrecy.
Malinformation is useful only if it makes that difficult middle more precise.
If the term becomes a shortcut for condemning uncomfortable truths, it has failed its own epistemic purpose.
Research references
- Wardle C, Derakhshan H. Information Disorder: Toward an interdisciplinary framework for research and policy making. Council of Europe. 2017. https://rm.coe.int/information-disorder-toward-an-interdisciplinary-framework-for-researc/168076277c
- Hameleers M. Disinformation as a Context-Bound Phenomenon. Communication Theory. 2023;33(1):1–10. Published online 2022. https://doi.org/10.1093/ct/qtac021
- Hayward M. What We Know About What They Know: An Epistemology of Disinformation. Social Epistemology. 2025. https://doi.org/10.1080/02691728.2024.2346127
- Pamirzad QH, Chen Q. The Conjuncture of Intentionality, Facticity, and Identity: Exposure to Disinformation and Malinformation on Social Media and Their Association With Ethnic Polarization. Social Media + Society. 2026. https://doi.org/10.1177/20563051261417516