Not all false or harmful information is the same. In 2017 the researchers Claire Wardle and Hossein Derakhshan, in a report for the Council of Europe, proposed three categories that are now widely used:

• Misinformation: false information shared without intent to cause harm. Your uncle forwards a fake health tip because he believes it and wants to help.
• Disinformation: false information deliberately created or shared to deceive or cause harm. A fake news site invents stories to earn advertising money or influence an election.
• Malinformation: genuine information shared to cause harm, such as publishing someone's private address or leaking real but private messages out of context to damage a person.

The same piece of content can move between categories. A fabricated story starts as disinformation, then becomes misinformation when sincere people pass it on.

Forms of false and misleading content

Wardle also described a range of types, from least to most deceptive:

• Satire or parody that is mistaken for real news.
• False connection: headlines or images that do not match the content.
• Misleading content: real information used to mislead.
• False context: genuine content shared with false information about when or where it happened, such as an old photo labeled as today's event.
• Imposter content: material that copies the name or logo of a real source.
• Manipulated content: genuine images or video altered to deceive.
• Fabricated content: entirely made up.

Why false claims spread

A large 2018 study of stories on Twitter from 2006 to 2017 found that false news spread farther, faster and more widely than true news. The researchers found that people, not automated bots, were mainly responsible, and suggested that false stories tend to be more novel and to provoke stronger emotions such as surprise and disgust.

Other reasons false claims spread:

• Repetition: the illusory truth effect, documented by psychologists since the 1970s, means that hearing a claim repeatedly makes it feel more true, even when we know better.
• Identity: claims that flatter our group or attack our opponents are easier to believe and share.
• Speed and ease: sharing takes one tap; checking takes effort.
• Money: sensational content attracts clicks, and clicks bring advertising revenue.

An old problem

False stories are not new. In 1835 the New York Sun published a series claiming that a famous astronomer had discovered winged creatures living on the moon, the so-called Great Moon Hoax, and its sales soared. Even stories about misinformation can be distorted: the 1938 radio drama The War of the Worlds is famous for causing mass panic, but later research suggests that, although some listeners were frightened, newspapers greatly exaggerated how many were fooled.

Edited media and deepfakes

Images have long been altered. In the Soviet Union, officials who fell from favor were removed from official photographs, as if they had never been there. Today, simple edits, such as slowing down a video to make a speaker seem drunk, cropping out context or mislabeling an old video, are sometimes called cheapfakes. They are common and effective because they require no advanced technology.

Deepfakes are images, audio or video generated or altered with artificial intelligence to show people saying or doing things they never did. The term appeared around 2017. As the technology improves, visual clues like strange blinking or blurred edges become less reliable, so checking the source and context matters more than inspecting pixels. Deepfakes also create a second problem, sometimes called the liar's dividend: when people know fakes exist, someone caught on a genuine recording can claim it is fake.

Worked example: the flood photo

During an invented hurricane, a photo spreads rapidly: a flooded city street with a large fish swimming past a submerged car. The caption reads: “This is downtown RIGHT NOW. Stay inside!!” It is shared tens of thousands of times.

A reverse image search shows the same photo appeared years earlier, during a different storm in a different country. The fish was added with editing software; the original photo showed only floodwater.

Classify what happened:

• The original edit was manipulated content, someone altering a genuine image.
• Re-posting it with today's date and location added false context.
• The person who first spread it this time may have known it was fake and wanted attention: disinformation.
• Most people sharing it sincerely wanted to warn friends: misinformation.

Why did it spread? It was novel and surprising, triggered fear during a frightening event, and took one tap to share. It also caused harm: people who later saw real flood warnings may have dismissed them as more hype.

Now compare a different case. During the same storm, someone posts the real home address of a city official, urging people to “go tell him what you think” about storm preparations. The address is accurate, so this is not false information. But sharing it to encourage harassment makes it malinformation. Accurate information can still cause harm when shared to target a person.

The lesson: in a crisis, when emotions run high and accurate information matters most, slow down before sharing, and stick to official emergency channels and established local news.