Are deepfakes a friend or a foe? That was the central question Pranit Biswas, managing associate at SS Rana & Co in New Delhi, asked the audience on the opening day of the Global IP Convention in Istanbul. “Most people have a negative view of deepfakes, but I hope to highlight some of the positive aspects of deepfakes,” he said.

Born out of machine learning and fake images, we have all been exposed to deepfakes, especially on social media. It can often be difficult to tell what is real or fake. What started harmless, with face-swapping and other light-hearted uses, deepfakes now can spread malicious content and be difficult to identify. In a recent Times of India survey, 86 percent of Indians thought that deepfakes could affect the outcome of an election, one of the more obvious negative ways the technology can be used to deceive.

“While the negative aspects cannot be ignored, AI and AI content, which is what deepfakes are, also have a lot of positive applications if they are used effectively,” said Biswas.

Effective use of deepfakes

In education, deepfake technology can effectively deliver engaging lessons and bring history alive. By recreating historical events and figures with realistic images, they can foster interactive learning experiences. Biswas highlighted their potential to improve learning outcomes through the use of synthetic voices and videos.

Biswas also illustrated a positive use of deepfake technology in healthcare, citing CereProc, a UK-based text-to-speech company, as an example. CereProc uses AI to create realistic deepfake voices for children and teenagers in Wales who cannot speak, granting them a voice reflecting who they are and their cultural background.

Deepfakes can also mask the identity of individuals where privacy is a concern, which is particularly useful for people living under repressive regimes or those seeking anonymity. Deepfake technology provides them with a realistic avatar and allows them to “speak” at conferences and online to protect their identities or privacy.