From Digital Footprints to AI Imprints: “Why the Age of AI Requires Us to Teach Both”

For more than two decades, one of the foundational messages in digital literacy education has been the concept of the digital footprint, or what we like to also call a digital dossier. Parents, caregivers, and educators have taught young people that what they do online can leave a trail. A photograph can be copied, a private conversation can be screenshot, a video can be downloaded, and a comment written in frustration can resurface years later. Something shared with a small group of friends can suddenly find its way into a much larger audience.

That lesson remains just as important today as it was when we first began talking about digital footprints. However, the rapid growth of artificial intelligence means we believe the conversation now needs to evolve.

Increasingly, it’s no longer enough to ask young people, “What am I leaving behind when I go online?” We also need to help them ask a second question, “What might my digital activity be helping to shape?” This is the distinction we make between a digital footprint and what we call an AI imprint.

A digital footprint is primarily about persistence, visibility, reputation, and the trail we leave behind. An AI imprint expands the conversation by recognizing that the photographs, videos, voices, words, searches, prompts, choices, interactions, and behaviours we generate online can potentially become part of the enormous pools of human generated information used to develop, improve, personalize, test, or influence the artificial intelligence systems increasingly surrounding us.

In other words, youth and teens aren’t simply growing up in a world where technology remembers what they do. They are increasingly growing up in a world where technology can learn from what people do.

Digital Footprints: What We Leave Behind

The digital footprint was one of the first major concepts used to help young people understand the consequences of participating online. The basic idea was relatively straightforward, when we interact with technology, we can leave traces of ourselves behind.

Those traces might include photographs, videos, comments, usernames, profiles, conversations, likes, searches, gaming activity, livestreams, posts, and countless other forms of digital information. Some are deliberately made public, while others are shared privately but can still be copied, forwarded, recorded, downloaded, or preserved by someone else.

This is why we have traditionally encouraged youth and teens to think before they post. The concern wasn’t simply what their friends might think today, but what could happen to that information tomorrow, next month, or years from now.

Could someone screenshot it? Could it be shared outside the intended audience? Could it affect a friendship? Could a university admissions officer, coach, scholarship committee, or future employer eventually discover it? Could something posted when they were fourteen become part of how someone evaluates them when they are twenty four?

The central digital footprint question became, “What am I leaving behind, and who might eventually see it?” That remains an important question. However, artificial intelligence introduces another dimension that the traditional digital footprint conversation wasn’t originally designed to address.

AI Imprints: What We Help Shape

Artificial intelligence systems don’t develop in isolation. They require information from which patterns can be identified, models can be tested, recommendations can be improved, and outputs can be generated. Much of the digital world already operates through machine learning systems that learn from enormous amounts of human behaviour and human created information. This is where the concept of an AI imprint becomes increasingly useful.

An AI imprint isn’t simply about whether something we post might still exist ten years from now. It asks us to consider whether our participation in digital environments can potentially influence the technological systems being built around us.

Every day, millions of people upload photographs, record videos, write comments, speak into microphones, search for information, watch particular content, skip other content, click links, follow accounts, interact with recommendations, communicate with AI systems, and generate enormous amounts of behavioural information.

Individually, many of these interactions can feel insignificant. Collectively, however, human activity provides valuable signals about language, culture, preferences, interests, relationships, attention, behaviour, creativity, communication, and countless other aspects of human life. This leads to a different digital literacy question, “What might my digital activity help teach, influence, or shape?” That’s the AI imprint.

Our Kids Are Creating More Than Content

Consider what an ordinary teenager might do during a typical week. They might upload a TikTok video, watch several hours of YouTube, comment on a friend’s Instagram post, participate in a Twitch livestream, search for information, send messages, upload photographs, create a gaming clip, interact with an AI chatbot, generate an AI image, ask an AI system to help with homework, and scroll through hundreds or even thousands of pieces of recommended content. From the teenager’s perspective, they are simply living their life. They are communicating, socializing, learning, creating, experimenting, playing, laughing, and being entertained.

From a technological perspective, however, many of those interactions can also generate information.

• What did they watch?

• What did they skip?

• What kept their attention?

• What did they search for?

• What did they create?

• What words did they use?

• What did they click?

• What did they ask an AI system?

• What response did they choose?

• What did they upload?

• What did they return to?

This doesn’t mean that every piece of information generated by every child is automatically being fed into a generative AI model. Platform practices, privacy rules, account settings, age, jurisdiction, and the particular service involved can all affect how information is collected and used. That distinction is important because the reality is more complicated than saying that every major social media platform simply has an “AI training switch” turned on by default - even though several do, such as Twitch.

For example, TikTok’s current privacy materials state that information it collects may be used to train, test, and improve technologies including machine-learning models and algorithms, including through analysis of user activity, content, messages, AI interactions, and associated metadata.  YouTube, however, currently says its separate setting allowing third-party companies to train AI models using eligible creator videos is off by default, illustrating why families shouldn’t assume every platform handles AI training in exactly the same way.

The larger lesson for families isn’t that every platform does exactly the same thing, it’s that we have entered an online environment where the information we generate can have value beyond the immediate reason we created it.

When Talking to AI Creates An Imprint Too

There is another part of this conversation that deserves particular attention, that being what young people directly provide to artificial intelligence systems.

When a youth or teenager interacts with an AI chatbot, they may feel as though they are having a private conversation with a computer. That perception can encourage people to disclose information they might never post publicly on social media.

A youth or teen might paste an assignment into an AI system. Another might upload a photograph and ask the AI to change it. Someone might provide information about a relationship problem, describe something happening at school, upload a document, record their voice, or provide details about themselves while asking the AI for advice. This creates an important new digital literacy lesson, “a prompt is also a form of disclosure.”

Before entering information into an AI system, young people should learn to ask many of the same questions we have taught them to ask before posting on social media. Do I need to provide this information? Does it identify me or someone else? Am I sharing somebody else’s private information? Does the service retain conversations? Can my interactions be reviewed by humans? Can they be used to improve the service or its machine-learning systems? Are there privacy settings available to me?

YouTube’s own privacy notice for its AI features, for example, specifically warns users not to include confidential or personal information when creating or interacting with those features. It also explains that some prompts, outputs, and related content may be reviewed to improve products and machine learning systems, while making distinctions about how certain recordings are treated. This is precisely why AI literacy now needs to become part of digital literacy.

The Difference Between a Footprint and an Imprint

A useful way to explain the distinction to a young person is this, “Your digital footprint is what technology may remember about you. Your AI imprint is what technology may learn from people like you.”

The footprint asks us to think about permanence, while the imprint asks us to think about influence. The footprint asks, “Where could this information go?” The imprint asks, “What could this information eventually help create, personalize, recommend, predict, or teach?” These concepts overlap, but they are not identical.

A youth or teenager’s photograph might become part of their digital footprint because someone screenshots and redistributes it. That same photograph, depending upon the platform, permissions, settings, and policies involved, could potentially have another technological use beyond simply being viewed by other people.

Similarly, watching a particular video may leave very little visible public footprint. However, the behavioural signal generated by watching it, replaying it, liking it, sharing it, or immediately scrolling past it can help recommendation systems understand what captures attention. The footprint is often visible, while the imprint can be largely invisible and that difference matters.

“But I Didn’t Agree to Teach an AI”

This is where the conversation becomes particularly important for parents and educators.

Most young people don’t join a social media platform because they want to contribute to the development of artificial intelligence, they join because their friends are there, they want entertainment, they want to create, they want to belong, and they want to watch videos, play games, share photographs, participate in communities, or communicate with others.

Similarly, a teenager using an AI chatbot usually isn’t thinking about data governance, machine learning, model development, licensing agreements, or privacy policies, they simply want an answer from their AI. However, the commercial digital environment is increasingly intertwined with artificial intelligence. Platforms are developing AI tools, recommendation technologies, generative features, automated moderation systems, advertising technologies, personalization systems, and other machine learning applications.

This doesn’t automatically make those technologies bad. Artificial intelligence can offer extraordinary opportunities for creativity, accessibility, education, productivity, communication, and problem solving. However, young people deserve to understand the exchange that can sometimes be taking place. The free service they are using isn’t necessarily free of value flowing in the opposite direction.

• Sometimes that value is attention.

• Sometimes it is advertising information.

• Sometimes it is behavioural data.

• Sometimes it is content.

• Sometimes it is feedback.

• Sometimes it may contribute to improving technological systems.

Digital literacy means understanding that relationship rather than simply fearing it.

Why “Don’t Post Anything You Wouldn’t Want Your Grandmother to See” Is No Longer Enough

For years, adults used sayings such as, “Don’t post anything you wouldn’t want your grandmother, teacher, or future employer to see.” Although well intentioned, that advice was always somewhat simplistic. Today, it is even less adequate.

A young person’s digital life is no longer limited to things they deliberately post publicly. Their digital presence can include what they watch, search, click, create, upload, prompt, generate, interact with, and sometimes even what they ignore.

Some of the most technologically valuable information a person generates may never appear publicly on their profile. This means digital literacy needs to move beyond simply teaching children to manage their online reputation. We also need to teach data awareness.

Before sharing something with a digital service, a useful question is no longer simply, “Would I be embarrassed if somebody saw this?” Another question should be, “Am I comfortable giving this information to this technology?” Those are very different questions.

This Shouldn’t Become Another Fear Based Conversation

Whenever a new technology emerges, there is a temptation to respond by frightening young people about everything that could possibly go wrong. We don’t believe that approach works, especially when it comes to AI.

Artificial intelligence will likely become even more deeply embedded in the educational, professional, creative, and social environments today’s children will eventually inherit. Telling young people simply to avoid AI isn’t preparing them for that world, teaching them how to use it thoughtfully is.

The goal should not be to convince children that every click is dangerous, every platform is stealing from them, or every AI interaction is somehow sinister. That would replace digital literacy with digital fear. Instead, we want young people to develop an informed habit of mind. Pause, think, and understand the exchange. Then decide.

• What am I giving this technology?

• Why does it need it?

• What might happen to it?

• Is there something identifying or sensitive in what I’m sharing?

• Do I understand the privacy settings?

• Could I accomplish the same goal while providing less information?

These are practical skills that travel with a young person from platform to platform and technology to technology.

Parents, Caregivers, and Educators Don’t Need to Become AI Engineers

One of the challenges facing adults right now is the feeling that artificial intelligence is evolving too quickly to keep up with. Parents, caregivers, and educators, don’t need to understand neural networks, large language model architecture, training pipelines, or every new AI product their child encounters. Educators don’t need to become computer scientists. What young people need from trusted adults is help developing the habit of asking good questions.

Instead of only asking, “What did you post?”, occasionally ask, “What information did you give the AI?” Instead of only discussing who can see a photograph, talk about what permissions an app has to access photographs. Instead of only teaching privacy as hiding information from strangers, teach privacy as making informed decisions about who and what receives our information, and why.

Most importantly, adults can model the same behaviour themselves. Before uploading a family photograph into an AI image generator, consider who appears in it. Before pasting a confidential work document into an AI chatbot, consider what information it contains. Before granting an app access to a photo library, microphone, contacts, or location, ask whether that access is actually necessary. Our children learn digital literacy not only from what we tell them, but from watching what we do.

From Digital Footprints to AI Imprints

The digital footprint isn’t an outdated concept, it’s the foundation upon which the next stage of digital literacy needs to be built. We still need to teach young people that information can persist, travel, be copied, become searchable, and affect their reputation. However, artificial intelligence adds another layer to that education.

Our children need to understand that their online lives don’t simply produce things that can be found later. Their participation can also produce information that may be used differently than they originally imagined.

That is the shift from thinking exclusively about digital footprints to also understanding AI imprints. Digital footprints teach young people to think about what they leave behind. AI imprints teach them to think about what they may be helping to shape, and both matter.

As artificial intelligence becomes increasingly integrated into the platforms, devices, classrooms, workplaces, and services surrounding our children, the goal shouldn’t be to make them afraid of participating in today’s onlife world. It should be to ensure they understand that participation has value, consequences, and increasingly, influence.

For the past twenty years, we have taught kids, “Think before you post.” Now the AI era requires us to add another lesson: 

“Think before you prompt, upload, click, share, or create, because your digital activity may not only become part of your story. It may also become part of what technology learns from humanity.”

Digital Food For Thought

The White Hatter

Facts Not Fear, Facts Not Emotions, Enlighten Not Frighten, Know Tech Not No Tech