Digital citizenship in the age of AI: Why checking before sharing matters more than ever

The ability to publish information to a potentially enormous audience was once the preserve of newspapers, broadcasters and other large organisations. Social media transformed that model, giving almost anyone the power to distribute photographs, video, commentary and news instantaneously.

Generative artificial intelligence is creating another shift. Images, voices, video and text that appear authentic can now be synthesised with relative ease. Against this background, verifying information before passing it on is becoming an important component of responsible digital citizenship.

The scale of the challenge is reflected in Canadian data. According to a recent Statistics Canada report highlighted by MediaSmarts, four in five Canadians reported encountering news or information online that they suspected was misleading, false or inaccurate at least monthly during the previous year. Nearly half reported that distinguishing true information from false information had become harder.

The figures provide the backdrop to Digital Citizen Day, which takes place on 28 October. The initiative, led by MediaSmarts, asks Canadians to consider the online environment they want and their own role in creating it. The event forms part of Media Literacy Week, running from 26 to 30 October.

Misinformation itself is hardly an invention of the artificial intelligence era. What generative AI changes is the speed and sophistication with which convincing material can be produced.

A realistic picture is no longer necessarily evidence that an event occurred. The same problem increasingly applies to audio and moving images. This pushes media literacy beyond the traditional question of whether a particular website appears trustworthy. People increasingly need to consider where an item originated, whether another reliable source corroborates it, and whether apparently compelling visual evidence actually demonstrates what somebody claims it does.

The developing problem has been a recurring theme in Digital Journal’s coverage. In September, Digital Journal reported on MediaSmarts’ Teen Fact-Checking Network, a bilingual programme through which Canadians aged 13 to 19 investigate viral claims and make fact-checking content aimed at their peers. Twenty-five teenage fact-checkers participated during the scheme’s first two years, with their videos attracting more than 7.5 million views.

That approach is significant because digital citizenship is not synonymous with simply avoiding questionable websites. It involves developing habits that can travel across platforms and technologies.

At a point when AI can manufacture increasingly persuasive digital material, the critical skill may not be recognising every deepfake on sight. It is knowing when the evidence in front of us is insufficient to justify clicking “share”.

Pause before amplifying

One particularly useful habit is deceptively simple: pause. Social networks make sharing frictionless. A striking photograph, extraordinary claim or alarming warning can move from one account to another within seconds. Yet precisely the material most likely to provoke an immediate emotional reaction can warrant additional scrutiny.

Verification can begin by looking beyond the post itself. Who originally made the claim? Does the link lead to the organisation or study being cited? Is the image being presented in its original context? Have reputable independent sources reported the same event?

For visual material, a reverse-image search can provide clues about whether an image appeared online previously and in a different context. MediaSmarts’ youth fact-checking programme specifically includes reverse-image searching among the verification techniques taught to participants.

It is an important distinction: verification does not necessarily mean proving personally that every component of a claim is true. It means looking for enough dependable evidence before deciding that the information deserves further distribution.

The stakes become higher when synthetic media enter the equation. Deepfakes can be used for impersonation and deception. Previous material researched for Digital Journal described concerns around increasingly realistic AI-generated video and the threat that manipulated media poses to identity and verification processes.

AI imagery also complicates assumptions about photographic evidence. Earlier Digital Journal research examined how generative image technology can produce convincing artificial material and how the availability of such technology creates new opportunities for manipulated evidence and impersonation.

Consequently, trying to identify a fake purely by looking for strange hands, distorted faces or other visual defects is unlikely to constitute a durable media-literacy strategy. As generation systems improve, yesterday’s tell-tale artefact may disappear. Provenance and corroboration therefore matter more. Someone seeing an extraordinary video should consider its source and seek independent reporting rather than assuming that apparent realism proves authenticity.

Teaching verification rather than fear

Young people are central to this discussion, but an effective response is not simply to tell them that the Internet is dangerous. MediaSmarts and the Canadian Internet Registration Authority recently launched Shields Up For Cyber Safety, a resource for students in Grades 7 to 9. The lesson asks pupils to act as “digital safety strategists”, identify risks associated with online activities and consider appropriate security tools and strategies. It uses what its developers describe as a strengths-first approach intended to build confidence, critical thinking and practical skills.

That educational model complements fact-checking initiatives because responsible citizenship requires both scepticism and agency. Treating every online claim as false would be no more useful than believing everything. The objective is to develop the ability to judge evidence. There is also a broader point behind Digital Citizen Day. Responsible behaviour online is about more than an individual’s ability to avoid being deceived. What people choose to share affects other users. Every unverified repost can potentially extend the reach of misleading information, while thoughtful participation can improve the quality of online discussion.

For 2026, therefore, good digital citizenship increasingly resembles a form of everyday information hygiene: check the origin, seek corroboration, examine context and pause before amplifying something uncertain.