Google is developing tools to fight AI image fraud in search
The new tools that will label AI-generated content in search results, helping users identify manipulated images and videos.
Recent surveys show that a majority of people are increasingly concerned about the threat of being deceived by deepfakes and the role AI could play in spreading propaganda.
These fears are not unfounded, given the number of online scams using AI-generated deepfakes that have surged over the past two years. A notable example occurred in February when a Hong Kong-based financier was tricked into transferring $25 million to scammers who impersonated the company's CFO during a video conference.
This rise in AI-related fraud is reflected in alarming statistics. According to report by Sumsub, a verification provider, scams involving AI-generated content increased by 245% between 2023 and 2024. Moreover, accounting firm Deloitte projects that financial losses tied to deepfakes will skyrocket from $12.3 billion in 2023 to $40 billion by 2027.
A significant factor contributing to this surge is the public accessibility of AI services, which has lowered the barrier for cybercriminals to exploit these tools.
The impact of AI-generated content isn't limited to scams alone. In recent months, AI-generated images have begun to dominate Google search results, making it harder for users to find accurate information.
Given the fact that Google processes 8.5 billion searches per day and approximately 2 trillion global searches per year according to SEO.AI, this has become a major issue.
Recognizing this, Google announced that it will start labelling AI-generated and AI-edited images in search results, aiming to improve transparency and user trust.
To implement this change, Google plans to use metadata provided by the Coalition for Content Provenance and Authenticity (C2PA), an industry group it joined earlier this year. The C2PA metadata will help track the origin of images, including when and where they were created, as well as the tools used in their generation. Companies like Amazon, Microsoft, OpenAI, and Adobe also support this initiative, showing a broader industry push toward content transparency.
Google will introduce these labels through an "About this image" feature, which will be integrated into Search, Google Lens, and Android’s Circle to Search functionality.
Additionally, the company plans to apply this technology to its ad services, with potential expansion to other platforms like YouTube in the future, though more details on that are expected later this year.
