Citron opens the keynote with the example of a journalist from India who was significantly affected by a viral deepfake video. The deepfake video was shared more than 40,000 times and led to harassment and the doxing of the journalist.
Danielle Citron explains that deepfake technology is a fabricated video or audio material created using machine learning. The content appears real, but is computer generated. Deepfake technology is widely available and improving at a rapid pace.
Citron outlines how vulnerable we are to this type of misinformation, and online hoaxes can spread much faster than accurate information. According to Danielle Citron, this can be partly attributed to confirmation bias, which occurs when individuals seek out information that matches a pre-existing viewpoint. Citron also states that social networks can often escalate issues related to confirmation bias.
Danielle Citron then outlines the implications that deepfakes can have for society. Deepfakes can be used to influence elections, or for corporate sabotage or other nefarious purposes. Furthermore, as technology improves, it may no longer be possible to distinguish a deepfake from an authentic video, and with rampant confirmation bias, Citron argues, we may begin to devalue the truth.
Finally, the keynote outlines the necessary actions to be taken. Laws must be passed to combat harassment, social media must remove harmful deepfake content, and society as a whole must adapt to emerging technology.
Related posts:
