“With every new technology advancement, there are new opportunities to help people — but also new forms of abuse that we need to combat.”
Google is amping up its efforts to remove explicit AI-generated deepfake images and videos from its search results.
The rise of generative AI image creation has resulted in an increase in the number of deepfakes, which have targeted social media users, celebrities, entrepreneurs and many others, necessitating stricter measures from tech giants to combat the growing threat. Thus, in a recent blog post, Google has outlined its plan to tackle the problem.
Easier ways to remove content:
Google has a history of allowing users to request the removal of deepfake content from their engine. However, they have now developed systems to make the process easier and applicable on a wider level.
According to Google: When someone successfully requests the removal of explicit non-consensual fake content featuring them from Search, Google’s systems will also aim to filter all explicit results on similar searches about them. In addition, when someone successfully removes an image from Search under our policies, our systems will scan for – and remove – any duplicates of that image that we find.”
Improved ranking systems:
Google will be launching ranking updates that will lower explicit fake content for many searches. For queries that are specifically seeking this content and include people’s names, they aim to surface non-explicit content, such as relevant news articles, when available. Thus, people will be able to read about the impact deepfakes have on society, rather than see pages with actual non-consensual fake images. For example, when searching for news about a celebrity’s deepfake and their testimony, Google will prioritize news stories and related articles over the deepfakes themselves.
Google is also making ongoing improvements to surface legitimate content and downrank explicit fake content. They are demoting sites that have received a high volume of removals for fake explicit imagery, an approach that has worked in other areas and which may prove valuable in reducing fake explicit content in search results.
According to Product Manager Emma Higham, these updates were made based on the feedback of rom experts and victim-survivors to ensure people’s protection. They have proven to be successful in addressing other types of non-consensual imagery and will hopefully combat fake explicit images and videos the same way.
However, the question remains: Will these updates actually work? Google reports some success in reducing explicit deepfakes, claiming that early tests with these changes have reduced the appearance of such images by over 70%. However, the search giant recognizes that more work is needed. Additionally, while Google’s updates may help in reducing the visibility of deepfakes on the internet, that does not change the existence of such content. For now, while mitigating the spread of harmful content is a positive step forward, there is still a great deal more work that needs to be done to tackle it entirely.
