By Lilian H. Hill
Artificial intelligence (AI) has dramatically changed how images are created and shared. Today, AI tools can generate highly realistic photographs, artwork, and videos from a simple text prompt within seconds. The image presented here is an example. While these technologies offer exciting opportunities for creativity, education, and communication, they also present serious risks when used to spread disinformation.
Misinformation refers to false or misleading information shared without the intent to deceive, whereas disinformation involves intentionally deceptive content designed to manipulate audiences. AI-generated images contribute to both forms of information disorder by creating visuals that appear authentic but are entirely fabricated. Researchers have warned that deepfakes and synthetic media increasingly blur the distinction between truth and fiction (Doss et al., 2023).
Humans Reliance on Visual Information
Images, whether realistic or AI-generated, are compelling because visual input plays a major role in how humans construct meaning. Vision is central to human cognition because it provides the brain with a vast amount of information about the environment, supporting perception, learning, memory, communication, and decision-making. Neuroscientists from MIT have found that the human brain can process entire images that the eye sees for as little as 13 milliseconds (Trafton, 2014).
One reason AI-generated images are particularly persuasive is their emotional impact. Humans often trust photographs as evidence of reality. Modern AI image generators now produce visuals with sophisticated lighting, facial detail, and contextual realism that make it difficult for viewers to identify false or deceptive images. According to Gil et al. (2023), advances in generative AI have significantly increased the realism and accessibility of deepfake technologies, making the creation of deceptive images easier than ever before.
The rapid spread of social media intensifies the problem. Platforms such as Facebook, Instagram, TikTok, and X prioritize emotionally engaging content, regardless of accuracy. False images connected to political scandals, natural disasters, wars, or public health crises can spread globally within minutes. Research on AI-generated misinformation has demonstrated that synthetic media is especially likely to go viral because emotionally provocative visuals encourage rapid sharing behaviors (Shoaib et al., 2023).
Several widely circulated examples illustrate the power of AI-generated deception. In 2023, an AI-generated image of Pope Francis wearing a white designer-style puffer coat spread rapidly online because many viewers believed it was authentic. Similarly, fabricated images of public figures depicted in tasteless or immoral activities circulate widely on social media before being debunked. These examples demonstrate how quickly synthetic images can shape public conversations and emotional reactions. Unfortunately, once shared, fabricated visuals can persist, influencing public perception and strengthening false narratives that are difficult to reverse (Katz, 2026).
Problems with AI-Generated Images
Political disinformation is one of the most concerning applications of AI-generated imagery. Fabricated images of politicians participating in criminal activity, attending false events, or endorsing fabricated positions can influence public opinion and undermine trust in democratic institutions. Łabuz and Nehring (2024) argued that deepfakes pose a growing threat to elections by manipulating voter perceptions and weakening confidence in democratic processes.
AI-generated images have transformed modern warfare into an era of synthetic information, shifting the battlefield into public perception and digital disinformation. World leaders now trade AI-generated images instead of negotiating differences privately. This synthetic media allows adversaries to rapidly manufacture fake evidence, including fabricated airstrikes, military engagements, and manipulated satellite imagery, to confuse populations and overwhelm the information environment. AI-generated images simulate events that never happened and are presented in ways that resemble authentic news generation, making them difficult to detect (Katz, 2026).
AI-generated misinformation also threatens public health communication. During health emergencies such as pandemics, fabricated images or videos can increase confusion, distrust, and panic. Experts have warned that deepfakes may intensify health misinformation by making fraudulent medical advice, fake experts, or fabricated emergencies appear credible to the public (Shoaib et al., 2023).
Further, AI-generated disinformation fuels scams and fraud by making deceptive content appear increasingly realistic and convincing. Fraudsters use AI to create fabricated images, videos, documents, and online identities that are difficult for many people to distinguish from authentic sources. These tactics often target members of the public with limited digital literacy and may lack the knowledge or confidence to evaluate the credibility of online content. As a result, individuals can be vulnerable to financial exploitation, identity theft, and other forms of deception.
Another significant concern is the erosion of trust in authentic media. As fake images become more common, individuals may begin doubting genuine photographs and videos as well. Jacobsen and Simpson (2024) described this phenomenon as the “liar’s dividend,” in which authentic evidence can be dismissed as fabricated because AI manipulation has become so plausible. The result is a broader crisis of credibility in journalism, science, and public communication.
Potential Solutions
Addressing these challenges requires multiple strategies. Technology companies are developing detection systems, digital watermarking tools, and authentication measures to identify synthetic media. Researchers continue exploring methods for detecting inconsistencies in AI-generated images, including visual artifacts and metadata analysis (Borji, 2023). However, detection technologies often struggle to keep pace with rapidly improving AI systems.
Media literacy education is equally important. The growing sophistication of AI-generated content underscores the need to strengthen digital literacy, critical media evaluation skills, and public awareness of emerging forms of online fraud. Verifying sources, checking multiple news outlets, using reverse image searches, and questioning emotionally provocative content are essential skills in the digital age. Educational initiatives emphasizing fact-checking and critical media consumption can help reduce the spread of misinformation and disinformation.
Developers and policymakers also face ethical responsibilities. Scholars have emphasized the need for stronger regulatory frameworks, transparent AI governance, and cross-platform collaboration to combat synthetic misinformation (Shoaib et al., 2023). Balancing innovation with accountability will remain one of the defining challenges of the AI era.
Conclusion
AI-generated images are not inherently harmful. They can support artistic creativity, accessibility, education, and innovation. Yet their misuse reveals how vulnerable modern societies are to visual deception. As AI becomes increasingly integrated into everyday communication, society must adapt to a new reality in which images alone can no longer be treated as unquestionable evidence. In the age of generative AI, seeing is no longer synonymous with believing.
References
Akhtar, Z. (2023). Deepfakes generation and detection: A short survey. Journal of Imaging, 9(1), 18. https://doi.org/10.3390/jimaging9010018
Borji, A. (2023). Qualitative failures of image generation models and their application in detecting deepfakes. Image and Vision Computing, 137, 104771. https://doi.org/10.1016/j.imavis.2023.104771
Doss, C., Mondschein, J., Shu, D., Wolfson, T., Kopecky, D., Fitton-Kane, V. A., Bush, L., & Tucker, C. (2023). Deepfakes and scientific knowledge dissemination. Scientific Reports, 13, 13429. https://doi.org/10.1038/s41598-023-39944-3
Gil, R., Virgili-Gomà, J., López-Gil, J.-M., & García, R. (2023). Deepfakes: Evolution and trends. Soft Computing, 27, 11295–11318. https://doi.org/10.1007/s00500-023-08605-y
Jacobsen, B. N., & Simpson, J. (2024). The tensions of deepfakes. Information, Communication & Society, 27(6), 1095–1109. https://doi.org/10.1080/1369118X.2023.2234980
Łabuz, M., & Nehring, C. (2024). On the way to deep fake democracy? Deep fakes in election campaigns in 2023. European Political Science, 23, 454–473. https://doi.org/10.1057/s41304-024-00482-9
Katz, D. (2026, March 30). Manufacturing reality: How AI generated war imagery enters the media supply chain. https://honestreporting.com/manufacturing-reality-how-ai-generated-war-imagery-enters-the-media-supply-chain/
Shoaib, M. R., Wang, Z., Ahvanooey, M. T., & Zhao, J. (2023). Deepfakes, misinformation, and disinformation in the era of frontier AI, generative AI, and large AI models. arXiv. https://doi.org/10.48550/arXiv.2311.17394
Trafton, A. (2014). In the blink of an eye. MIT News. https://news.mit.edu/2014/in-the-blink-of-an-eye-0116
