Flipping the script on deepfake technologies
September 21, 2021
September 21, 2021
Companies view deepfakes as cause for attention and concern, and it’s easy to see why. Almost anyone can access deepfake technology to make a convincing video of someone saying things they never said.
But this focus on negative implications has led most companies to overlook a new opportunity. Ultimately, these technologies are about creating realistic but synthetic data that can have real value. There are applications for industries ranging from entertainment to education to health and life sciences. With the proper technological and ethical approaches, synthetic data capabilities can drive significant business value. And today, much of that value is still untapped.
At its core, generative network technology – the tech behind deepfakes – is about creating realistic synthetic data. In the case of a deepfake, the goal is to create audio or video that can fool human viewers. But synthetic data can generate value in many enterprise applications, across many industries.
Labs has used synthetic data to speed up product development while exploring more possible formulations than before.
Synthetic data is used to train machine learning systems when there isn’t enough real data available, or it isn’t practical to obtain.
Deepfake technologies can support customized content for more inclusive communication and wider reach.
Companies can use synthetic data sets for modelling and analysis without exposing real data.
Synthetic data can generate value in many enterprise applications, across many industries
Like every technology, generative approaches create risks for business and society. When companies apply them to address business challenges and opportunities, they must do so responsibly. To maintain trust with consumers, regulators, and the public, companies must incorporate ethical considerations into their decision-making process from the start.
Every technological innovation has potential for abuse. Given how heavily our societies and business models rely on information, deepfakes perhaps have more potential for misuse than most. It’s shortsighted, though, to focus only on the potential for negative impact – and businesses doing so are leaving value on the table.
In everything from product development to healthcare to the entertainment industry, synthetic data offers real value. How will you capture it?