Artificial intelligence and elections

From Wikipedia, the free encyclopedia

https://en.wikipedia.org/wiki/Artificial_intelligence_and_elections 

A screenshot of a deepfake video from June 2023 showing an AI-generated debate between Donald Trump and Joe Biden taken from the Twitch channel TrumporBiden2024

As artificial intelligence (AI) has become more mainstream, there is growing concern about how this will influence elections. Potential targets of AI include election processes, election offices, election officials and election vendors. There are also global efforts to improve elections using AI.

Tactics

Generative AI capabilities allow creation of misleading content. Examples of this include text-to-video, deepfake videos, text-to-image, AI-altered images, text-to-speech, voice cloning, and text-to-text. In the context of an election, a deepfake video of a candidate may propagate information that the candidate does not endorse. Chatbots could spread misinformation related to election locations, times or voting methods. In contrast to malicious actors in the past, these techniques require little technical skill and can spread rapidly.

LLM-generated messages have the capacity to persuade humans on political issues. Researchers have begun to investigate how people rate messages that LLMs generate for how persuasive they are. When it came to policy issues, the LLM-generated messages received a 2.91 compared to a 2.80 when it came to smartness between the AI and humans. The LLM-generated messages were often more technical and analytical than human-generated messages.

Generative AI has been used to micro-target people during tight political elections. The generation of targeted large language models has triggered concern that they will be used to leverage readily scale microtargeting. Rephrasing inputs have been used to generate fraudulent emails and phishing websites. Rephrasing inputs in a microtargeting does not violate the terms of OpenAI usage. There are no safeguards to prevent the use of rephrasing and creation of fraudulent emails. Political campaign managers have access to this allowing for them to create targeted content.

Usage by country

Argentina

2023 elections

During the 2023 Argentine primary elections, Javier Milei's team distributed AI generated images including a fabricated image of his rival Sergio Massa and drew 3 million views. The team also created an unofficial Instagram account entitled "AI for the Homeland." Sergio Massa's team also distributed AI generated images and videos.

Bangladesh

2024 elections

In the run up to the 2024 Bangladeshi general election, deepfake videos of female opposition politicians appeared. Rumin Farhana was pictured in a bikini while Nipun Ray was shown in a swimming pool.

Canada

2025 elections

In the run up to the 2025 Canadian federal election, the use of AI tools is likely to figure prominently. India, Pakistan and Iran are all expected to make efforts to subvert the national vote using disinformation campaigns to deceive voters and sway diaspora communities.

In a report by the Canadian Centre for Cyber Security called "Cyber Threats to Canada's Democratic Process: 2025 Update", it states that malicious actors including China and Russia: "are most likely to use generative AI as a means of creating and spreading disinformation, designed to sow division among Canadians and push narratives conducive to the interests of foreign states".

France

2024 elections

In the 2024 French legislative election, deepfake videos appeared claiming: i) That they showed the family of Marine le Pen. In the videos, young women, supposedly Le Pen's nieces, are seen skiing, dancing and at the beach "while making fun of France’s racial minorities": However, the family members don't exist. On social media there were over 2 million views. ii) In a video seen on social media, a deepfake video of a France24 broadcast appeared to report that the Ukrainian leadership had "tried to lure French president Emmanuel Macron to Ukraine to assassinate him and then blame his death on Russia".

Ghana

2024 elections

During the months before the December 2024 Ghanaian general election, a network of at least 171 fake accounts has been used to spam social media. Posts have been used by a group identified as "@TheTPatriots" to promote the New Patriotic Party, although it is not known whether the two are connected. All the networks' posts were "highly likely" to have been generated by ChatGPT and appear to be the "first secretly partisan network using AI to influence elections in Ghana". The opposition National Democratic Congress was also criticized with its leader John Mahama being called a drunkard.

India

2024 elections

In the 2024 Indian general election, politicians used deepfakes in their campaign materials. These deepfakes included politicians who had died prior to the election. Mathuvel Karunanidhi's party posted with his likeness even though he had died 2018. A video The All-India Anna Dravidian Progressive Federation party posted showed an audio clip of Jayaram Jayalalithaa even though she had died in 2016. The Deepfakes Analysis Unit (DAU) is an open source platform created in March 2024 for the public to share misleading content and assess if it had been AI-generated.

AI was also used to translate political speeches in real time. This translating ability was widely used to reach more voters.

Indonesia

2024 elections

In the 2024 Indonesian presidential election, Prabowo Subianto made extensive use of AI-generated art in his campaign, which ranged from images of himself as an adorable child to various child portrayals in his advertisements. The Indonesian Children's Protection Commission condemned these ads, labeling them as a form of misuse. Other candidates, Anies Baswedan and Ganjar Pranowo, also incorporated AI art into their campaigns. Throughout the election period, all presidential candidates faced attacks from deepfakes, both in video and audio formats.

Ireland

2024 elections

In the last weeks of the 2024 Irish general election a spoof election poster appeared in Dublin featuring "an AI-generated candidate with three arms". The candidate is called Aidan Irwin, but no-one stood in the election with that name. A slogan on the poster says "put matters into artificial intelligence’s hands". The convincing election poster shows a man that "has six fingers on one hand, three arms, and a distorted thumb".

New Zealand

2023 elections

In May 2023, ahead of the 2023 New Zealand general election in October 2023, the New Zealand National Party published a "series of AI-generated political advertisements" on its Instagram account. After confirming that the images were faked, a party spokesperson said that it was "an innovative way to drive our social media".

Pakistan

2024 elections

AI has been used by the imprisoned ex-Prime Minister Imran Khan and his media team in the 2024 Pakistani general election: i) An AI generated audio of his voice was added to a video clip and was broadcast at a virtual rally. ii) An op-ed in The Economist written by Khan was later claimed by himself to have been written by AI which was later denied by his team. The article was liked and shared on social media by thousands of users.

South Africa

2024 elections

In the 2024 South African general election, there were several uses of AI content: i) A deepfaked video of Joe Biden emerged on social media showing him saying that "The U.S. would place sanctions on SA and declare it an enemy state if the African National Congress (ANC) won". ii) In a deepfake video, Donald Trump was shown endorsing the uMkhonto weSizwe party. It was posted to social media and was viewed more than 158,000 times. iii) Less than 3 months before the elections, a deepfake video showed U.S. rapper Eminem endorsing the Economic Freedom Fighters party while criticizing the ANC. The deepfake was viewed on social media more than 173,000 times.

South Korea

2022 elections

In the 2022 South Korean presidential election, a committee for one presidential candidate Yoon Suk Yeol released an AI avatar 'Al Yoon Seok-yeol' that would campaign in places the candidate could not go. The other presidential candidate Lee Jae-myung introduced a chatbot that provided information about the candidate's pledges.

2024 elections

Deepfakes were used to spread misinformation before the 2024 South Korean legislative election with one source reporting 129 deepfake violations of election laws within a two week period.

Seoul hosted the 2024 Summit for Democracy, a virtual gathering of world leaders initiated by US President Joe Biden in 2021. The focus of the summit was on digital threats to democracy including artificial intelligence and deepfakes.

Taiwan

2024 elections

AI-generated content was used during the 2024 Taiwanese presidential election. Among the media were: i) A deepfake video of General Secretary of the Chinese Communist Party Xi Jinping which showed him supporting the presidential elections. Created on social media, the video was "widely circulated" and often "accompanied by claims that Xi supported candidates from one of the two opposition parties". ii) In a deepfake video U.S. congressman Rob Wittman is shown appearing to support Taiwan's Democratic Progressive Party. The video shows him saying that the U.S. would increase its military support, accelerating "all arms sales to Taiwan." It was shown on various social media platforms.

United Kingdom

2024 elections

The Centre for Emerging Technology and Security provided a report on the threat of AI to the 2024 UK general election. The reports' findings said that the impact of AI was limited but may damage the democratic system.

In the run up to the UK 2024 general elections, AI-generated videos spread extensively on social media including: i) A deepfake video showed then PM Rishi Sunak claiming that he would "require 18-year-olds to be sent to active war zones in Gaza and Ukraine as part of their national service". The video had more than 40,000 views. ii) A deepfake video showed PM Keir Starmer "swearing repeatedly at a staffer". Comments from the original poster included calling Starmer a "disgusting bully". The social media site showing the video refused to delete it despite requests.

Entrepreneur Steve Endacott from the south of England created "AI Steve," an AI avatar as the face of his campaign for member of parliament.

United States

2024 elections

Main article: Artificial intelligence in the 2024 United States presidential election

Further information: Foreign interference in the 2024 United States elections

Officials from the ODNI and FBI have stated that Russia, Iran, and China used generative artificial intelligence tools to create fake and divisive text, photos, video, and audio content to foster anti-Americanism and engage in covert influence campaigns. The use of artificial intelligence was described as an accelerant rather than a revolutionary change to influence efforts. Regulation of AI with regard to elections was unlikely to see a resolution for most of the 2024 United States general election season.

The campaign for the 2024 Republican nominee, Donald Trump, has used deepfake videos of political opponents in campaign ads and fake images showing Trump with black supporters. In 2023, while he was still running for re-election, the presidential campaign of Joe Biden prepared a task force to respond to AI images and videos.

A Democratic consultant working for Dean Phillips also admitted to using AI to generate a robocall which used Joe Biden's voice to discourage voter participation.

Generative AI increased the efficiency with which political candidates were able to raise money by analyzing donor data and identifying possible donors and target audiences.

Regulation

See also: Regulation of artificial intelligence

By governments

Philippines

The Commission on Elections (COMELEC) issued guidelines on the usage of AI, to be implemented starting from the 2025 Philippine general election including the Bangsamoro Parliament election originally scheduled for 2025. It mandates candidate to disclose usage of AI in their campaign materials and prohibits the usage of the technology to spread misinformation against their rivals. This is the first time the COMELEC has release guidelines on campaigning through social media.

United States

US states have attempted regulation of AI use in elections and campaigns with varying degrees of success. The National Conference of State Legislatures has compiled a list of legislation regarding AI use by state as of 2024, some carrying both civil and criminal penalties. Oregon Senate Bill 1571 requires that campaign communications in Oregon disclose the use of AI. California has enacted legislation that makes using deepfakes to discredit political opponents illegal within sixty days of an election.

Self-regulation by private firms

Midjourney, an AI image-generator, has started blocking users from creating fake images of the 2024 US Presidential candidates. Research from the Center for Countering Digital Hate found that image generators such as Midjourney, ChatGPT Plus, DreamStudio, and Microsoft's Image Creator create images that constitute election disinformation in 41% of the test text prompts they tried. OpenAI implemented policies to counter election misinformation such as adding digital credentials to image origin and a classifier to detect if images were AI generated.

AI use in election interference by foreign governments

See also: Foreign electoral intervention

AI has begun to be used in election interference by foreign governments. Governments thought to be using AI to interfere in external elections include Russia, Iran and China. Russia was thought to be the most prolific nation targeting the 2024 presidential election with their influencing operations "spreading synthetic images, video, audio and text online", according to U.S intelligence officials. Iran has reportedly generated fake social media posts stories and targeted "across the political spectrum on polarizing issues during the presidential election". The Chinese government has used "broader influence operations" that aim to make a global image and "amplify divisive topics in the U.S. such as drug use, immigration, and abortion". For example, Spamouflage has increasingly used generative AI for influence operations.

Outside of the US elections, a deepfake video of Moldova’s pro-Western president Maia Sandu shows her "throwing her support behind a political party friendly to Russia." Officials in Moldova "believe the Russian government is behind the activity". Slovakia's liberal party leader had audio clips faked which discussed "vote rigging and raising the price of beer". The Chinese government has used AI to stir concerns about US interference in Taiwan. A fake clip seen on social media showed a fake video of the vice chairman of the U.S. House Armed Services Committee promising "stronger U.S. military support for Taiwan if the incumbent party’s candidates were elected in January".

Ethics of AI use in political campaigning

See also: Ethics of artificial intelligence

As the use of AI and its associated tools in political campaigning and messaging increases, many ethical concerns have been raised. Campaigns have used AI in a number of ways, including speech writing, fundraising, voter behaviour prediction, fake robocalls and the generation of fake news. At the moment there are no US federal rules when it comes to using AI in campaigning and so its use can undermine public trust. Yet according to one expert: "A lot of the questions we're asking about AI are the same questions we've asked about rhetoric and persuasion for thousands of years."

As more insight into how AI is used becomes ever greater, concerns have become much broader than just the generating of misinformation or fake news. Its use by politicians and political parties for "purposes that are not overtly malicious" can also raise ethical worries. For instance, the use of 'softfakes' have become more common. These can be images, videos or audio clips that have been edited, often by campaign teams, "to make a political candidate seem more appealing." An example can be found in Indonesia's presidential election where the winning candidate created and promoted cartoonish avatars so as to rebrand himself.

How citizens come by information has been increasingly impacted by AI, especially through online platforms and social media. These platforms are part of complex and opaque systems which can result in a "significant impact on freedom of expression", with the generalisation of AI in campaigns also creating huge pressures on "voters’ mental security". As the frequency of AI use in political campaigning becomes common, together with globalization, more 'universalized' content can be used so that territorial boundaries matter less. While AI collides with the reasoning processes of people, the creation of "dangerous behaviours" can happen which disrupt important levels of society and nation states.

Catch-22 (logic)

From Wikipedia, the free encyclopedia

https://en.wikipedia.org/wiki/Catch-22_(logic) 

A flowchart showing Joseph Heller's original catch-22

A catch-22 is a paradoxical situation from which an individual cannot escape because of contradictory rules or limitations. The term was first used by Joseph Heller in his 1961 novel Catch-22.

Catch-22s often result from rules, regulations, or procedures that an individual is subject to, but has no control over, because to fight the rule is to accept it. Another example is a situation in which someone is in need of something that can only be had by not being in need of it (e.g., the only way to qualify for a loan is to prove to the bank that you do not need a loan). One connotation of the term is that the creators of the "catch-22" situation have created arbitrary rules in order to justify and conceal their own abuse of power.

Origin and meaning

Joseph Heller coined the term in his 1961 novel Catch-22, which describes absurd bureaucratic constraints on soldiers in World War II. The term is introduced by the character Doc Daneeka, an army surgeon who invokes "Catch-22" to explain why any pilot requesting mental evaluation for insanity—hoping to be found not sane enough to fly and thereby escape dangerous missions—demonstrates his own sanity in creating the request and thus cannot be declared insane. This phrase also means a dilemma or difficult circumstance from which there is no escape because of mutually conflicting or dependent conditions.

"You mean there's a catch?"

"Sure there's a catch," Doc Daneeka replied. "Catch-22. Anyone who wants to get out of combat duty isn't really crazy."

There was only one catch and that was Catch-22, which specified that a concern for one's own safety in the face of dangers that were real and immediate was the process of a rational mind. Orr was crazy and could be grounded. All he had to do was ask; and as soon as he did, he would no longer be crazy and would have to fly more missions. Orr would be crazy to fly more missions and sane if he didn't, but if he was sane, he had to fly them. If he flew them, he was crazy and didn't have to; but if he didn't want to, he was sane and had to. Yossarian was moved very deeply by the absolute simplicity of this clause of Catch-22 and let out a respectful whistle.

Different formulations of "Catch-22" appear in the novel. The term is applied to loopholes and quirks of the military system, always with the implication that rules are inaccessible to and slanted against those lower in the hierarchy. In chapter 6, Yossarian is told that Catch-22 requires him to do anything his commanding officer tells him to do, regardless of whether these orders contradict orders from the officer's superiors.

In a final episode, Catch-22 is described to Yossarian by an old woman recounting an act of violence by soldiers,

"Catch-22 says they have a right to do anything we can't stop them from doing."

"What the hell are you talking about?" Yossarian shouted at her in bewildered, furious protest. "How did you know it was Catch-22? Who the hell told you it was Catch-22?"

"The soldiers with the hard white hats and clubs. The girls were crying. 'Did we do anything wrong?' they said. The men said no and pushed them away out the door with the ends of their clubs. 'Then why are you chasing us out?' the girls said. 'Catch-22,' the men said. All they kept saying was 'Catch-22, Catch-22.' What does it mean, Catch-22? What is Catch-22?"

"Didn't they show it to you?" Yossarian demanded, stamping about in anger and distress. "Didn't you even make them read it?"

"They don't have to show us Catch-22," the old woman answered. "The law says they don't have to."

"What law says they don't have to?"

"Catch-22."

According to literature professor Ian Gregson, the old woman's narrative defines "Catch-22" more directly as the "brutal operation of power", stripping away the "bogus sophistication" of the earlier scenarios.

Other appearances in the novel

Besides referring to an unsolvable logical dilemma, Catch-22 is invoked to explain or justify the military bureaucracy. In the first chapter, it requires Yossarian to sign his name to letters he censors while he is confined to a hospital bed. One clause mentioned in chapter 10 closes a loophole in promotions, which one private had been exploiting to re-attain the attractive rank of private first class after any promotion. Through courts-martial for going AWOL, he would be busted back to private but Catch-22 limited the number of times he could do this before being sent to the stockade.

At another point in the book, a prostitute explains to Yossarian that she cannot marry him because he is crazy, and she will never marry a crazy man. She considers any man crazy who would marry a woman who is not a virgin. This closed logic loop clearly illustrated Catch-22 because by her logic, all men who refuse to marry her are sane and thus she would consider marriage; but as soon as a man agrees to marry her, he becomes crazy for wanting to marry a non-virgin, and is instantly rejected.

Captain Black attempts to press Milo into depriving Major Major of food as a consequence of not signing a loyalty oath that Major Major was never given an opportunity to sign in the first place. Captain Black asks Milo, "You're not against Catch-22, are you?"

In chapter 40, Catch-22 forces Colonels Korn and Cathcart to promote Yossarian to Major and ground him rather than simply sending him home. They fear that if they do not, others will refuse to fly, just as Yossarian did.

Significance of the number 22

Main articles: Catch-22 § Title, and Catch-22

Heller originally wanted to call the phrase (and hence, the book) by other numbers, but he and his publishers eventually settled on 22. The number has no particular significance; it was chosen more or less for euphony. The title was originally Catch-18, but Heller changed it after the popular Mila 18 was published a short time beforehand.

Usage

The term "catch-22" has filtered into common usage in the English language. In a 1975 interview, Heller said the term would not translate well into other languages.

James E. Combs and Dan D. Nimmo suggest that the idea of a "catch-22" has gained popular currency because so many people in modern society are exposed to frustrating bureaucratic logic. They write of the rules of high school and colleges that:

This bogus democracy that can be overruled by arbitrary fiat is perhaps a citizen's first encounter with organizations that may profess 'open' and libertarian values, but in fact are closed and hierarchical systems. Catch-22 is an organizational assumption, an unwritten law of informal power that exempts the organization from responsibility and accountability, and puts the individual in the absurd position of being excepted for the convenience or unknown purposes of the organization.^[5]

Along with George Orwell's "doublethink", "Catch-22" has become one of the best-recognized ways to describe the predicament of being trapped by contradictory rules.^[9]

A significant type of definition of alternative medicine has been termed a Catch-22. In a 1998 editorial co-authored by Marcia Angell, a former editor of the New England Journal of Medicine, argued that:

It is time for the scientific community to stop giving alternative medicine a free ride. There cannot be two kinds of medicine—conventional and alternative. There is only medicine that has been adequately tested and medicine that has not, medicine that works and medicine that may or may not work. Once a treatment has been tested rigorously, it no longer matters whether it was considered alternative at the outset. If it is found to be reasonably safe and effective, it will be accepted. But assertions, speculation, and testimonials do not substitute for evidence. Alternative treatments should be subjected to scientific testing no less rigorous than that required for conventional treatments.

This definition has been described by Robert L. Park as a logical Catch-22 which ensures that any complementary and alternative medicine (CAM) method which is proven to work "would no longer be CAM, it would simply be medicine."

U.S. Circuit Judge Don Willett referred to qualified immunity, which requires a violation of constitutional rights to have been previously established in order for a victim to claim damages, as a Catch-22: "Section 1983 meets Catch-22. Important constitutional questions go unanswered precisely because those questions are yet unanswered. Courts then rely on that judicial silence to conclude there's no equivalent case on the books. No precedent = no clearly established law = no liability. An Escherian Stairwell. Heads government wins, tails plaintiff loses."

Logic

The archetypal catch-22, as formulated by Joseph Heller, involves the case of John Yossarian, a U.S. Army Air Forces bombardier, who wishes to be grounded from combat flight. This will only happen if he is evaluated by the squadron's flight surgeon and found "unfit to fly". "Unfit" would be any pilot who is willing to fly such dangerous missions, as one would have to be mad to volunteer for possible death. To be evaluated, he must request the evaluation, an act that is considered sufficient proof for being declared sane. These conditions make it impossible to be declared "unfit".

The "Catch-22" is that "anyone who wants to get out of combat duty isn't really crazy". Pilots who request a mental fitness evaluation are sane, and therefore must fly in combat. At the same time, if an evaluation is not requested by the pilot, he will never receive one and thus can never be found insane, meaning he must also fly in combat. Catch-22 ensures that no pilot can ever be grounded for being insane even if he is.

A logical formulation of this situation is:

1.	${\displaystyle (E\to (I\wedge R))}$	For a person to be excused from flying on the grounds of insanity (E), he must both be insane (I) and have requested an evaluation (R).	(premise)
2.	${\displaystyle (I\to \mathrm{\neg }R)}$	An insane person (I) does not request an evaluation (¬R) because he does not realize he is insane.	(premise)
3.	${\displaystyle (\mathrm{\neg }I\vee \mathrm{\neg }R)}$	Either a person is not insane (¬I) or does not request an evaluation (¬R).	(2. and material implication)
4.	${\displaystyle (\mathrm{\neg }(I\wedge R))}$	No person can be both insane (I) and request an evaluation (R).	(3. and De Morgan's laws)
5.	${\displaystyle (\mathrm{\neg }E)}$	Therefore, no person can be excused from flying on the grounds of insanity (¬E) because no person can be both insane and have requested an evaluation.	(4., 1. and modus tollens)

The philosopher Laurence Goldstein argues that the "airman's dilemma" is logically not even a condition that is true under no circumstances; it is a "vacuous biconditional" that is ultimately meaningless. Goldstein writes:

The catch is this: what looks like a statement of the conditions under which an airman can be excused flying dangerous missions reduces not to the statement

(i) 'An airman can be excused flying dangerous missions if and only if Cont' (where 'Cont' is a contradiction)

(which could be a mean way of disguising an unpleasant truth), but to the worthlessly empty announcement

(ii) 'An airman can be excused flying dangerous missions if and only if it is not the case that an airman can be excused flying dangerous missions'

If the catch were (i), that would not be so bad—an airman would at least be able to discover that under no circumstances could he avoid combat duty. But Catch-22 is worse—a welter of words that amounts to nothing; it is without content, it conveys no information at all.

Disinformation

From Wikipedia, the free encyclopedia

https://en.wikipedia.org/wiki/Disinformation

Disinformation is false or misleading information deliberately spread to deceive people, or to secure economic or political gain and which may cause public harm. Disinformation is an orchestrated adversarial activity in which actors employ strategic deceptions and media manipulation tactics to advance political, military, or commercial goals. Disinformation is implemented through coordinated campaigns that "weaponize multiple rhetorical strategies and forms of knowing—including not only falsehoods but also truths, half-truths, and value judgements—to exploit and amplify culture wars and other identity-driven controversies."

In contrast, misinformation refers to inaccuracies that stem from inadvertent error. Misinformation can be used to create disinformation when known misinformation is purposefully and intentionally disseminated. "Fake news" has sometimes been categorized as a type of disinformation, although scholars distinguish between the two concepts. Some researchers have cautioned against using the term in academic writing have because of its lack of precision and its contested use in public and political discourse regarding news coverage or information. Others note that it continues to be used both in reference to demonstrably false or misleading information and more broadly in political communication, where its meaning and implications remain debated.

Etymology

The Etymology of Disinformation by H. Newman as published in The Journal of Information Warfare in 2021. Elements of the word disinformation have their origins in Proto-Indo-European language family. The Latin 'dis' and 'in' and can both be considered to have Proto-Indo-European roots, 'forma' is considerably more obscure. The green box in the figure highlights the origin 'forma' is uncertain, however, it may have its roots in the Aristotelian concept of μορφή (morphe) where something becomes a 'thing' when it has 'form' or substance.

The English word disinformation comes from the application of the Latin prefix dis- to information making the meaning "reversal or removal of information". The rarely used word had appeared with this usage in print at least as far back as 1887. Some consider it a loan translation of the Russian дезинформация, transliterated as dezinformatsiya, apparently derived from the title of a KGB black propaganda department. Soviet planners in the 1950s defined disinformation as "dissemination (in the press, on the radio, etc.) of false reports intended to mislead public opinion."

Disinformation first made an appearance in dictionaries in 1985, specifically, Webster's New College Dictionary and the American Heritage Dictionary. In 1986, the term disinformation was not defined in Webster's New World Thesaurus or New Encyclopædia Britannica. After the Soviet term became widely known in the 1980s, native speakers of English broadened the term as "any government communication (either overt or covert) containing intentionally false and misleading material, often combined selectively with true information, which seeks to mislead and manipulate either elites or a mass audience."

By 1990, use of the term disinformation had fully established itself in the English language within the lexicon of politics. By 2001, the term disinformation had come to be known as simply a more civil phrase for saying someone was lying. Stanley B. Cunningham wrote in his 2002 book The Idea of Propaganda that disinformation had become pervasively used as a synonym for propaganda.

Operationalization

The Shorenstein Center at Harvard University defines disinformation research as an academic field that studies "the spread and impacts of misinformation, disinformation, and media manipulation", including "how it spreads through online and offline channels, and why people are susceptible to believing bad information, and successful strategies for mitigating its impact". According to a 2023 research article published in New Media & Society, disinformation circulates on social media through deception campaigns implemented in multiple ways including: astroturfing, conspiracy theories, clickbait, culture wars, echo chambers, hoaxes, fake news, propaganda, pseudoscience, and rumors.

Glossary of key terms related to disinformation and misinformation
Term	Description	Term	Description
Astroturfing	A centrally coordinated campaign that mimics grassroots activism by making participants pretend to be ordinary citizens	Fake news	Genre: The deliberate creation of pseudo-journalism Label: The instrumentalization of the term to delegitimize news media
Conspiracy theories	Rebuttals of official accounts that propose alternative explanations in which individuals or groups act in secret	Greenwashing	Deceptive communication makes people believe that a company is environmentally responsible when it is not
Clickbait	The deliberate use of misleading headlines and thumbnails to increase online traffic for profit or popularity	Propaganda	Organized mass communication, on a hidden agenda, and with a mission to conform belief and action by circumventing individual reasoning
Culture wars	A phenomenon in which multiple groups of people, who hold entrenched values, attempt to steer public policy contentiously	Pseudoscience	Accounts that claim the explanatory power of science, borrow its language and legitimacy but diverge substantially from its quality criteria
Doxxing	A form of online harassment that breaches privacy boundaries by releasing information intending physical and online harm to a target	Rumors	Unsubstantiated news stories that circulate while not corroborated or validated
Echo chamber	An epistemic environment in which participants encounter beliefs and opinions that coincide with their own	Trolling	Networked groups of digital influencers that operate 'click armies' designed to mobilize public sentiment
Hoax	News in which false facts are presented as legitimate	Urban legends	Moral tales featuring durable stories of intruders incurring boundary transgressions and their dire consequences

In order to distinguish between similar terms, including misinformation and malinformation, scholars collectively agree on the definitions for each term as follows: (1) disinformation is the strategic dissemination of false information with the intention to cause public harm; (2) misinformation represents the unintentional spread of false information; and (3) malinformation is factual information disseminated with the intention to cause harm, these terms are abbreviated 'DMMI'.

In 2019, Camille François devised the "ABC" framework of understanding different modalities of online disinformation:

• Manipulative Actors, who "engage knowingly and with clear intent in viral deception campaigns" that are "covert, designed to obfuscate the identity and intent of the actor orchestrating them." Examples include personas such as Guccifer 2.0, Internet trolls, state media, and military operatives.
• Deceptive Behavior, which "encompasses the variety of techniques viral deception actors may use to enhance and exaggerate the reach, virality and impact of their campaigns." Examples include troll farms, Internet bots, astroturfing, and "paid engagement".
• Harmful Content, which includes health misinformation, manipulated media such as deepfakes, online harassment, violent extremism, hate speech or terrorism.

In 2020, the Brookings Institution proposed amending this framework to include Distribution, defined by the "technical protocols that enable, constrain, and shape user behavior in a virtual space". Similarly, the Carnegie Endowment for International Peace proposed adding Degree ("distribution of the content ... and the audiences it reaches") and Effect ("how much of a threat a given case poses").

Comparisons with propaganda

Whether and to what degree disinformation and propaganda overlap is subject to debate. Some (like U.S. Department of State) define propaganda as the use of non-rational arguments to either advance or undermine a political ideal, and use disinformation as an alternative name for undermining propaganda, while others consider them to be separate concepts altogether. One popular distinction holds that disinformation also describes politically motivated messaging designed explicitly to engender public cynicism, uncertainty, apathy, distrust, and paranoia, all of which disincentivize citizen engagement and mobilization for social or political change.

Practice

Disinformation is the label often given to foreign information manipulation and interference (FIMI). Studies on disinformation are often concerned with the content of activity whereas the broader concept of FIMI is more concerned with the "behaviour of an actor" that is described through the military doctrine concept of tactics, techniques, and procedures (TTPs).

Disinformation is primarily carried out by government intelligence agencies, but has also been used by non-governmental organizations and businesses. Front groups are a form of disinformation, as they mislead the public about their true objectives and who their controllers are. Most recently, disinformation has been deliberately spread through social media in the form of "fake news", disinformation masked as legitimate news articles and meant to mislead readers or viewers. Disinformation may include distribution of forged documents, manuscripts, and photographs, or spreading dangerous rumours and fabricated intelligence. Use of these tactics can lead to blowback, however, causing such unintended consequences such as defamation lawsuits or damage to the dis-informer's reputation.

Disinformation can spread with greater ease in situations where there is a lack of credible information on a topic, such as a crisis. The implications of this were seen after the 2024 Southport stabbings, where the spread of disinformation regarding the killer's background was fuelled by reporting restrictions, meaning that there were no facts to counter the disinformation. This then led to the 2024 United Kingdom riots.

Worldwide

Disinformation has been identified as a significant challenge, undermining public trust and the factual basis for public debate. Studies have found that false information often spreads faster and more widely online than accurate information. Empirical research has also links disinformation to political polarisation, democratic erosion, and regime resilience in authoritarian contexts.

Soviet disinformation

This section is an excerpt from Soviet disinformation.edit

Former Romanian secret police senior official Ion Mihai Pacepa exposed disinformation history in his book Disinformation (2013).

Use of disinformation as a Soviet tactical weapon started in 1923, when it became a tactic used in the Soviet political warfare called active measures.

Russian disinformation

Russian disinformation campaigns have occurred in many countries. For example, disinformation campaigns led by Yevgeny V. Prigozhin, a Russian oligarch, have been reported in several African countries. Russia denies that it uses disinformation to influence public opinion. Often Russian campaigns aim to disrupt domestic politics within Europe and the United States in an attempt to weaken the West due to its long-standing commitment to fight back against "Western imperialism" and shift the balance of world power to Russia and her allies (the Second World during the Cold War era). According to the Voice of America, Russia seeks to promote American isolationism, border security concerns and racial tensions within the United States through its disinformation campaigns.

Chinese disinformation

Spamouflage (also known under the names Dragonbridge, Spamouflage Dragon, Storm 1376 and Taizi Flood) is an online propaganda and disinformation operation that has been using a network of social media accounts to make posts in favor of the Chinese Communist Party and government of the People's Republic of China and harass dissidents and journalists overseas since 2017. Beginning in the early 2020s, Spamouflage accounts also began making posts about American and Taiwanese politics. It is widely believed that the Chinese government, particularly the Ministry of Public Security, is behind the network. Spamouflage has increasingly used generative artificial intelligence for influence operations. The campaign has largely failed to receive views from real users, although it has attracted some organic engagement using new tactics.

American disinformation

How Disinformation Can Be Spread, explanation by U.S. Defense Department (2001)

The United States Intelligence Community appropriated use of the term disinformation in the 1950s from the Russian dezinformatsiya, and began to use similar strategies, starting during Cold War, and in conflict with other nations. The New York Times reported in 2000 that during the CIA's effort to substitute Mohammed Reza Pahlavi for then-Prime Minister of Iran Mohammad Mossadegh, the CIA placed fictitious stories in the local newspaper. Reuters documented how, subsequent to the 1979 Soviet Union invasion of Afghanistan during the Soviet–Afghan War, the CIA put false articles in newspapers of Islamic-majority countries, inaccurately stating that Soviet embassies had "invasion day celebrations". Reuters noted a former U.S. intelligence officer said they would attempt to gain the confidence of reporters and use them as secret agents, to affect a nation's politics by way of their local media.

In October 1986, the term gained increased currency in the U.S. when it was revealed that two months previously, the Reagan Administration had engaged in a disinformation campaign against then-leader of Libya Muammar Gaddafi. White House representative Larry Speakes said reports of a planned attack on Libya as first broken by The Wall Street Journal on August 25, 1986, were "authoritative", and other newspapers including The Washington Post then wrote articles saying this was factual. U.S. State Department representative Bernard Kalb resigned from his position in protest over the disinformation campaign, and said: "Faith in the word of America is the pulse beat of our democracy."

The executive branch of the Reagan administration kept watch on disinformation campaigns through three yearly publications by the Department of State: Active Measures: A Report on the Substance and Process of Anti-U.S. Disinformation and Propaganda Campaigns (1986); Report on Active Measures and Propaganda, 1986–87 (1987); and Report on Active Measures and Propaganda, 1987–88 (1989).

According to a report by Reuters, the United States ran a propaganda campaign to spread disinformation about the Sinovac Chinese COVID-19 vaccine, including using fake social media accounts to spread the disinformation that the Sinovac vaccine contained pork-derived ingredients and was therefore haram under Islamic law. Reuters said the ChinaAngVirus disinformation campaign was designed to "counter what it perceived as China's growing influence in the Philippines" and was prompted by the "[fear] that China's COVID diplomacy and propaganda could draw other Southeast Asian countries, such as Cambodia and Malaysia, closer to Beijing". The campaign was also described as "payback for Beijing's efforts to blame Washington for the pandemic". The campaign primarily targeted people in the Philippines and used a social media hashtag for "China is the virus" in Tagalog. The campaign ran from 2020 to mid-2021. The primary contractor for the U.S. military on the project was General Dynamics IT, which received $493 million for its role.

Since 2023, Republican members of the US Congress have attacked researchers who study disinformation as being against freedom of speech and as a euphemism for government censorship. On April 18, 2025, citing an Executive Order signed by Trump, the US National Science Foundation released a statement cancelling funding for disinformation research, saying that it does not fit with the NSF priorities, "including but not limited to those on diversity, equity, and inclusion (DEI) and misinformation/disinformation."

Disinformation in the 2016 United Kingdom European Union membership referendum

The 2016 United Kingdom European Union membership referendum saw figures from the Leave campaign spread disinformation on social media. Many of these arguments played on voters' emotions and attachments, such as the claim that the £350 million that the UK sent to the EU each week could be allocated to the NHS instead, which made them highly effective. Ofcom requires news broadcasters to remain neutral, yet this adherence to neutrality created false balance in the Brexit debate, causing broadcasters to inadvertedly promote disinformation.

Response

Responses from cultural leaders

Pope Francis condemned disinformation in a 2016 interview, after being made the subject of a fake news website during the 2016 U.S. election cycle which falsely claimed that he supported Donald Trump. He stated that the worst thing the news media could do was spread disinformation and said the act was a sin, comparing those who spread disinformation to individuals who engage in coprophilia.

Ethics in warfare

In a contribution to the 2014 book Military Ethics and Emerging Technologies, writers David Danks and Joseph H. Danks discuss the ethical implications in using disinformation as a tactic during information warfare. They note there has been a significant degree of philosophical debate over the issue as related to the ethics of war and use of the technique. The writers describe a position whereby the use of disinformation is occasionally allowed, but not in all situations. Typically the ethical test to consider is whether the disinformation was performed out of a motivation of good faith and acceptable according to the rules of war. By this test, the tactic during World War II of putting fake inflatable tanks in visible locations on the Pacific Islands in order to falsely present the impression that there were larger military forces present would be considered as ethically permissible. Conversely, disguising a munitions plant as a healthcare facility in order to avoid attack would be outside the bounds of acceptable use of disinformation during war.

Research

Main article: Disinformation research

Disinformation spreads through controversies.

Research related to disinformation studies is increasing as an applied area of inquiry. The call to formally classify disinformation as a cybersecurity threat is made by advocates due to its increase in social networking sites. Despite the proliferation of social media websites, Facebook and Twitter showed the most activity in terms of active disinformation campaigns. Techniques reported on included the use of bots to amplify hate speech, the illegal harvesting of data, and paid trolls to harass and threaten journalists.

Whereas disinformation research focuses primarily on how actors orchestrate deceptions on social media, primarily via fake news, new research investigates how people take what started as deceptions and circulate them as their personal views. As a result, research shows that disinformation can be conceptualized as a program that encourages engagement in oppositional fantasies (i.e., culture wars), through which disinformation circulates as rhetorical ammunition for never-ending arguments. As disinformation entangles with culture wars, identity-driven controversies constitute a vehicle through which disinformation disseminates on social media. This means that disinformation thrives, not despite raucous grudges but because of them. The reason is that controversies provide fertile ground for never-ending debates that solidify points of view.

Scholars have pointed out that disinformation is not only a foreign threat as domestic purveyors of disinformation are also leveraging traditional media outlets such as newspapers, radio stations, and television news media to disseminate false information. Current research suggests right-wing online political activists in the United States may be more likely to use disinformation as a strategy and tactic. The 2016 European Union referendum in the UK also saw British politicians supporting the Leave campaign spread disinformation on Twitter. The Conservative Party was also accused of spreading disinformation about the Labour Party in the lead-up to the 2019 General Election. Governments have responded with a wide range of policies to address concerns about the potential threats that disinformation poses to democracy, however, there is little agreement in elite policy discourse or academic literature as to what it means for disinformation to threaten democracy, and how different policies might help to counter its negative implications.

Consequences of exposure to disinformation online

There is a broad consensus amongst scholars that there is a high degree of disinformation, misinformation, and propaganda online; however, it is unclear to what extent such disinformation has on political attitudes in the public and, therefore, political outcomes. This conventional wisdom has come mostly from investigative journalists, with a particular rise during the 2016 U.S. election: some of the earliest work came from Craig Silverman at Buzzfeed News. Cass Sunstein supported this in #Republic, arguing that the internet would become rife with echo chambers and informational cascades of misinformation leading to a highly polarized and ill-informed society. Later studies have since proven the existence of echo chambers on social media.

Research after the 2016 US presidential election found: (1) for 14% of Americans social media was their "most important" source of election news; 2) known false news stories "favoring Trump were shared a total of 30 million times on Facebook, while those favoring Clinton were shared 8 million times"; 3) the average American adult saw fake news stories, "with just over half of those who recalled seeing them believing them"; and 4) people are more likely to "believe stories that favor their preferred candidate, especially if they have ideologically segregated social media networks." Correspondingly, whilst there is wide agreement that the digital spread and uptake of disinformation during the 2016 election was massive and very likely facilitated by foreign agents, there is an ongoing debate on whether all this had any actual effect on the election. For example, a double blind randomized-control experiment by researchers from the London School of Economics (LSE), found that exposure to online fake news about either Trump or Clinton had no significant effect on intentions to vote for those candidates. Researchers who examined the influence of Russian disinformation on Twitter during the 2016 US presidential campaign found that exposure to disinformation was (1) concentrated among a tiny group of users, (2) primarily among Republicans, and (3) eclipsed by exposure to legitimate political news media and politicians. Finally, they find "no evidence of a meaningful relationship between exposure to the Russian foreign influence campaign and changes in attitudes, polarization, or voting behavior." As such, despite its mass dissemination during the 2016 Presidential Elections, online fake news or disinformation probably did not cost Hillary Clinton the votes needed to secure the presidency.

Research on this topic remains inconclusive, for example, misinformation appears not to significantly change political knowledge of those exposed to it. There seems to be a higher level of diversity of news sources that users are exposed to on Facebook and Twitter than conventional wisdom would dictate, as well as a higher frequency of cross-spectrum discussion. Other evidence has found that disinformation campaigns rarely succeed in altering the foreign policies of the targeted states.

Research is also challenging because disinformation is meant to be difficult to detect and some social media companies have discouraged outside research efforts. For example, researchers found disinformation made "existing detection algorithms from traditional news media ineffective or not applicable...[because disinformation] is intentionally written to mislead readers...[and] users' social engagements with fake news produce data that is big, incomplete, unstructured, and noisy." Facebook, the largest social media company, has been criticized by analytical journalists and scholars for preventing outside research of disinformation.

Alternative perspectives and critiques

Researchers have criticized the framing of disinformation as being limited to technology platforms, removed from its wider political context and inaccurately implying that the media landscape was otherwise well-functioning. "The field possesses a simplistic understanding of the effects of media technologies; overemphasizes platforms and underemphasizes politics; focuses too much on the United States and Anglocentric analysis; has a shallow understanding of political culture and culture in general; lacks analysis of race, class, gender, and sexuality as well as status, inequality, social structure, and power; has a thin understanding of journalistic processes; and, has progressed more through the exigencies of grant funding than the development of theory and empirical findings."

Alternative perspectives have been proposed:

1. Moving beyond fact-checking and media literacy to study a pervasive phenomenon as something that involves more than news consumption.
2. Moving beyond technical solutions including AI-enhanced fact checking to understand the systemic basis of disinformation.
3. Develop a theory that goes beyond Americentrism to develop a global perspective, understand cultural imperialism and Third World dependency on Western news, and understand disinformation in the Global South.
4. Develop market-oriented disinformation research that examines the financial incentives and business models that nudge content creators and digital platforms to circulate disinformation online.
5. Include a multidisciplinary approach, involving history, political economy, ethnic studies, feminist studies, and science and technology studies.
6. Develop understandings of Gendered-based disinformation (GBD) defined as "the dissemination of false or misleading information attacking women (especially political leaders, journalists and public figures), basing the attack on their identity as women."

Strategies for spreading disinformation

Disinformation attack

Main article: Disinformation attack

The research literature on how disinformation spreads is growing. Studies show that disinformation spread in social media can be classified into two broad stages: seeding and echoing. "Seeding", when malicious actors strategically insert deceptions, like fake news, into a social media ecosystem, and "echoing" is when the audience disseminates disinformation argumentatively as their own opinions often by incorporating disinformation into a confrontational fantasy.

Internet manipulation

Internet manipulation is the use of online digital technologies, including algorithms, social bots, and automated scripts, for commercial, social, military, or political purposes. Internet and social media manipulation are the prime vehicles for spreading disinformation due to the importance of digital platforms for media consumption and everyday communication. When employed for political purposes, internet manipulation may be used to steer public opinion, polarise citizens, circulate conspiracy theories, and silence political dissidents. Internet manipulation can also be done for profit, for instance, to harm corporate or political adversaries and improve brand reputation. Internet manipulation is sometimes also used to describe the selective enforcement of Internet censorship or selective violations of net neutrality.

Internet manipulation for propaganda purposes with the help of data analysis and internet bots in social media is called computational propaganda.

Studies show four main methods of seeding disinformation online:

1. Selective censorship
2. Manipulation of search rankings
3. Hacking and releasing
4. Directly sharing disinformation

Recently, concerns have been expressed that AI technology could enable groups of AI programs, known as 'AI swarms,' to autonomously coordinate to manipulate democratic discourse, by maintaining persistent fake identities and mimicking human social dynamics over extended periods of time.

Exploiting online advertising technologies

Disinformation is amplified online due to malpractice concerning online advertising, especially the machine-to-machine interactions of real-time bidding systems. Online advertising technologies have been used to amplify disinformation due to the financial incentives and monetization of user-generated content and fake news. The lax oversight over the online advertising market can be used to amplify disinformation, including the use of dark money used for political advertising.