How to Spot AI Images and Videos on Social Media
Your guide to recognising AI-generated content
AI-generated content is rapidly spreading across the internet and social media. While often harmless, it can be used to mislead, be deceitful and even cause harm. Here's what to watch out for and why learning to recognise AI-made images and videos matters.
What these AI-generated images are designed to do is provoke an immediate reaction. Whether it is shock, anger or compassion, these images are triggering your emotions even before you have a chance to ask yourself whether they are real.
If it seems impossible, it probably is. Before sharing content online, take a moment to trust your common sense, especially when something looks too dramatic or too perfect to be real.
Things to be on the lookout for:
- Images are often accompanied with exaggerated, misspelled or alarming headlines, designed to grab your attention and encourage you to click and share.
- Check to see if some parts of the image are blurred or pixelated.
- The audio doesn’t match the movement of the lips or the scene, and the voices or voice overs are distorted or sound off or robotic.
- The protagonists in the background appear strange with overly intense expressions or none at all. Or human-like emotions on animals (for example, a 'crying' dog, a cat looking wildly terrified).
- The text shown doesn’t make sense.
- Look for distorted or added body parts. Unnatural poses, or rigid, robotic movements in videos.
- Do the protagonist's shadows align naturally? AI-generated ones often don't.
- Check the post's hashtags. Sometimes the AI will be mentioned there.
- Check other posts from the account.
- Look for subtle watermarks.
- AI sometimes struggles with letters and numbers. Zoom in on details such as license plates, logos, signs, etc.
How to ensure trustable sources and avoid scams
Scams involving fake shelters or rescues often rely on AI-generated images and vague information to provoke strong emotions and push people to donate. Be cautious if the page provides little to no real details, no location, no rescue information, no phone number, and only a suspicious donation link or requests to share your number privately. Legitimate organisations can provide proof of payment or official documents. If you're not 100% sure where your money is going, don't donate!
AI-generated scam content also tends to show animals in extremely exaggerated or unrealistic distress to manipulate viewers, so keep a critical eye before donating or even sharing posts, as spreading false content can harm others too.
Also, look at the account itself. Is it new? Does it have any history? What happens in the comments? If people are flagging AI use or asking for details and their comments get deleted or they get blocked, that's a strong sign the account is not trustworthy.
Scammers depend on engagement to go viral, so don't fuel it and stop the cycle
Posts that trigger strong emotions tend to get more reactions, comments, and shares. The more people interact with them, the more social media algorithms push them onto everyone's feeds. This is exactly what scammers and click-hungry accounts rely on. Whether they are after donations or just trying to go viral, high engagement works in their favour.
How these posts affect animals in real life
AI-generated content is increasingly shaping how people think (wild) animals behave and often creating unrealistic and dangerous expectations. For example, making interactions with wild animals and making it seem easy or harmless, or portraying them doing tricks, dances, or even human habits. These AI-generated content are not funny and goes against FOUR PAWS' principles of using animals for entertainment.
At the same time, fake dramatic rescue videos and fabricated shelters undermine the hard work of legitimate organisations. It steals the attention and the support from real animals and real charities that truly need help. These trends only mislead the public and put both humans and animals at risk. By spreading manipulated or staged AI content, it is harming the credibility of real organisations doing important work.
How to report and flag to not feed the algorithm
If you spot misleading or harmful AI content, don't interact with it. Report it as a scam, fraud, or false information so it stops spreading. Avoid boosting the post by commenting or resharing. If you want to warn others take a screenshot instead sharing it. The people behind these posts are chasing views, and every interaction, even a negative one, helps them. So please report them, it is the best way to stop the cycle.
