Deepfakes threaten upcoming elections, but 'responsible AI' could help filter them out before they reach us

The advancements in generative AI have significantly impacted the line between fact and fiction in media. With the ability to create realistic deepfakes - manipulated videos, images, or audio that appear authentic but are not - the distinction between truth and falsehood has become increasingly blurred. This poses a critical threat to the democratic process, as deepfakes can sow seeds of distrust in the political landscape and potentially sway election outcomes.

Generative AI has made it easier and less expensive to create deepfakes, making it accessible to those who may want to influence elections or spread disinformation. As a result, the challenge of distinguishing fact from fiction has become even more difficult, and the consequences can be far-reaching.

To combat this issue, the concept of "responsible AI" has emerged, where technology is developed and used in a way that it conforms to ethical principles and standards. This approach aims to detect and filter out deepfakes, much like a spam filter, preventing them from reaching the public. Additionally, governments and tech companies are exploring regulatory measures, such as mandatory disclosure of AI use in election adverts and incorporating watermarks in AI content, to tackle the problem of deepfakes and disinformation.

The US is taking several measures to regulate AI and deepfakes. One significant action is the launch of an AI taskforce to explore ways to effectively regulate AI and deepfakes, especially in the context of political campaigns and elections. This taskforce aims to find solutions that can prevent the harmful impacts of deepfakes and misinformation on democratic processes.

In addition, federal and state lawmakers have proposed legislation to regulate AI and deepfakes. At the federal level, there have been proposals to specifically target the use of deepfakes and other manipulated content in federal elections. At the state level, laws banning or restricting deepfakes and other deceptive media in election advertisements and political messages have been passed in several states, including California, Minnesota, Texas, and Washington.

On the other hand, India plans to introduce penalties for both the creators of deepfakes and other forms of disinformation, and the platforms that spread it. This approach holds not only the individuals who create the deepfakes accountable, but also the platforms that facilitate their distribution.

Both the US and India are actively working on finding effective ways to regulate AI and deepfakes to prevent their potential harm to society, particularly in the context of democratic processes.