Artificial intelligence-created material has become ubiquitous online, creating growing challenges for people trying to distinguish authentic information from fabricated content, especially during breaking news events.
Recent conflicts have highlighted this problem dramatically. Following the February 28 attacks by the United States and Israel on Iran, researchers documented an extraordinary volume of deceptive and fraudulent images created through artificial intelligence that circulated to millions worldwide. These included fabricated bombing footage depicting events that never occurred, pictures of allegedly captured military personnel, and Iranian propaganda materials showing President Donald Trump and other figures as blocky, Lego-style characters.
The 10th annual International Fact-Checking Day, observed today, offers an ideal moment to examine these growing concerns.
False information produced through AI technology spreads at remarkable rates across countless platforms. Throughout the Iran conflict, social media accounts representing various perspectives in the dispute have promoted such fabricated material.
The Institute for Strategic Dialogue, an organization that monitors false information and online extremism, has analyzed social media activity surrounding the Iran conflict. Their research revealed approximately two dozen X platform accounts that consistently share AI-created content and accumulated over one billion views combined since hostilities began. Many of these accounts displayed blue verification checkmarks.
Below are strategies for identifying AI-produced content versus authentic material in a digital environment where this distinction becomes increasingly challenging.
Early AI-generated imagery often contained clear indicators revealing their artificial nature. People might appear with incorrect numbers of fingers, audio might not match lip movements, or text could be meaningless. Objects frequently appeared warped or lacked essential elements. While advancing technology has reduced these obvious signs, they remain worth checking. Look for inconsistencies like vehicles appearing and disappearing in videos or actions defying physical laws. Some images may appear excessively refined or possess an artificial glossiness.
AI-created images circulate repeatedly across platforms. Tracing their origins helps determine authenticity. Conducting reverse image searches provides a straightforward method for this investigation. For videos, capture a screenshot first. This process might lead to social media profiles specializing in AI content creation, older images being misused, or completely unexpected discoveries.
Seek multiple credible sources to verify image authenticity. This might include fact-checking reports from established news organizations, official statements from public officials, or posts from misinformation specialists. These sources often possess sophisticated AI detection methods or access to information unavailable to general users.
Numerous AI detection applications can provide useful starting points. However, exercise caution since these tools don’t always provide accurate evaluations. Content created or modified through Google’s Gemini application includes invisible digital watermarking technology called SynthID, which the application can identify. Other AI creation platforms add visible watermarks to their generated material. These marks are typically easy to eliminate, so their absence doesn’t guarantee image authenticity.
Sometimes fundamental approaches work best. Pause, breathe deeply, and avoid immediately sharing unverified content. Malicious actors often exploit people’s emotional responses and preexisting beliefs when reacting to material. Reading comment sections may reveal whether images are genuine. Other users might spot details you missed or locate original sources. Remember that determining with complete certainty whether images are AI-generated isn’t always feasible, so stay aware they might be fabricated.
