Despite coming under waves of criticism of late, Facebook is perhaps the one social media platform leading the charge against deepfakes, having recently setup the Deepfake Detection Challenge in a bid to better identify manipulated media, which is becoming a greater concern as the US presidential elections near.
Joining Facebook in its efforts is Amazon, or more specifically Amazon Web Services (AWS), which is now a technical partner for the initiative, as well as being a member of the committee that will oversee the Challenge.
Another element that AWS is adding is up to $1 million in credits for any team that requires cloud computing in order to complete relevant tasks associated with the Challenge – namely trying to detect a deepfake.
The Challenge officially launches in December, which will also see a dataset of 100 000 created samples being introduced for AI-designed applications to try to spot the deepfakes. Leading up to this date Facebook is making an initial 5 000 samples available, as teams registering for the Challenge begin testing out the capabilities of their solution.
With the likes of Microsoft having already joined Facebook for the Deepfake Detection Challenge, adding AWS to the mix should add a bit more clout, and how that all parties concerned are taking the threat of deepfakes seriously.
While it will be interesting to see how effective the deepfake detecting solutions are at their task, it still remains to be seen what Facebook plans to do with the AI-designed projects following the Challenge.
Could one of the teams create a solution that will be included on Facebook to help identify deepfakes in future?
We will have to wait and see, with the Challenge running until March next year.