Think you can still spot AI? Here’s how to catch more convincing AI images, deepfakes, and fraud

AI images, videos, and text used to be relatively easy to find. There were too many fingers, distorted backgrounds, strange and nonsensical text and other visual artifacts, and it quickly stopped being an AI game. Deepfake videos often had delayed lip-syncing, and the AI-generated text felt repetitive and formulaic, like reading a clunky LinkedIn post.

But AI has evolved. Many images, videos, and messages generated by AI now appear convincing enough to fool even the most careful observer or expert. This means that some of the old tricks for telling what is and isn’t AI don’t work as reliably as they once did.

So this is not about finding signs or glitches. It’s about taking extra steps to confirm what you’re seeing and hearing. You can no longer rely on zooming in to see six fingers on a weird looking ad. You need to know what questions to ask and what tools to use when something doesn’t feel right.

from the latest videos

Fact-checking your Facebook page

Facebook, Messenger, WhatsApp, Instagram, Oculus apps, smartphone silhouette with blurred META logo on background

(Image credit: Shutterstock / mundissima)

AI-generated content abounds on Facebook, especially photos and videos designed to evoke strong emotional responses.

A Full Fact investigation in May looked at pages sharing AI-generated articles about British politicians and found that many were controlled from outside the UK, despite using British-sounding names. The posts featured heartwarming stories of politicians donating millions of dollars, rescuing dogs, and helping sick children. None of that was true.

One tool that can help in this situation is Facebook’s Page Transparency feature. This information within a Page’s profile can reveal where the Page is maintained from, when it was created, whether it has been renamed in the past, and other clues about who is actually behind the Page. If these details don’t match the image your page is trying to project, that’s often enough reason to be skeptical.

Spotting scams on LinkedIn

Zoom example seen on LinkedIn

(Image credit: LinkedIn)

Recruitment scams aren’t new, but with the advent of AI, they’re becoming harder to detect. Scammers can now generate realistic recruiter messages, professional profiles, and convincing emails at scale.

Some recent campaigns have mimicked LinkedIn notifications and job alerts, using urgency and curiosity to pressure people into clicking malicious links or sharing sensitive information.

Before accepting an unexpected job offer, do some basic checks. Does the recruiter have a connection to a real company? Does their profile look legitimate? Is the business present at Companies House? Taking a few minutes to review opportunities can save you a lot of trouble later on.

Viral video verification

A person using a cell phone trapped in a scrolling hole surrounded by a collage of social media obsessions.

(Image credit: Getty Images / We Are)

Deepfake videos are becoming eerily realistic. At normal speeds, AI-generated videos may look completely real. However, increasing the playback speed makes it easier to spot subtle discrepancies. Pay attention to lip movements that don’t quite match the speech, unnatural eye blinks, strange pacing, or facial movements that feel slightly out of sync with the audio you’re hearing.

Also pay attention to the facial expressions. Of course, no one reacts perfectly all the time, but if the person’s facial expressions consistently seem to diverge from what they’re saying, it may be worth investigating further.

AI influencer research

AI actor Tilly Norwood sits in her office and stares at the camera.

(Image credit: Xicoia)

Sometimes the answer is hidden in plain sight. You just need to know where to look. AI influencers, AI-assisted fashion shoots, and AI-generated ad campaigns are often published by brands and creators, but that information can be hard to find in captions, hashtags, profile descriptions, or small print. For example, lifestyle publisher and brand SheerLuxe creates AI-generated content and influencers under the name Sheerluxe lab.

Also try a reverse image search. When you upload an image to Google Images, we may see where the image first appeared online, whether it’s been modified, and whether the people in the image actually exist elsewhere on the internet. If an influencer only appears in AI-generated content and nowhere else, that’s a pretty big sign that it could be a complete hoax.