Deepfakes should be treated 'differently' from other online misinformation, says Zuckerberg

But Facebook CEO offers no immediate solutions to the disinformation problem

Facebook CEO Mark Zuckerberg has admitted that the company has yet to come up with a strategy for dealing with deepfake videos.

Speaking at an event in Aspen, Colorado, on Wednesday, Zuckerberg said: "We are going through the policy process of thinking through what the deepfake policy should be," adding: "the policies continue to evolve as technology develops".

Deepfake videos, footage altered by AI to make it appear that the person in the video is doing or saying something that they are not, have already been used for the purpose of spreading political disinformation. An example is the recent video featuring US House Speaker Nancy Pelosi slurring her words during a speech as if drunk, which as the original footage shows, never actually happened. Facebook was slow to remove the video due to an apparent "execution mistake". This led to strong complaints from several Democrats, with more than 25 sending Zuckerberg a letter saying they were "concerned that you and your company are not taking these occurrences seriously and are grossly unprepared for the 2020 elections."

A deepfake video of Zuckerberg was posted to Instagram, which Facebook owns, on 11 June. The video was posted not for the purpose of misinformation, but in order to provoke a response from Zuckerberg and to see if a video featuring Zuckerberg himself would be taken down more quickly. However, the video was not taken down and remains on the site.

Zuckerberg alluded to concerns over perceived censorship on the site, saying, cryptically, "I do not think we want to go so far towards saying that a private company prevents you from saying something that it thinks is factually incorrect to another person".

The main concern regarding deepfake videos is that fakes will soon be indistinguishable from the real thing, further blurring the boundaries between fact and fiction, and undermining trust in all reportage. Although AI is a massively innovative field with many potential positives, there are also disturbing implications that the technology could be used for disinformation, blackmail, framing, or potentially even to start a war.

Zuckerberg said that deepfakes may need to be treated in a different way from other forms of misinformation, which many believe Facebook has failed to manage properly.

"It is likely sensible to have a different policy and to treat this differently than how we just treat normal, false information on the internet," he said.