Deepfake 'News Anchors' Appear In Pro-China Footage On Social Media, Research Group Says
A deep fake 'newsanchors' appear in pro-china footage. A new research report says that people with ties to the Chinese government are using AI-made "deep fake news" broadcasters to spread pro-China propaganda videos on social mediaplatforms.
In its report, the US-based research firm Graphika said that the fake anchors for a fake news outlet called Wolf News were made by artificial intelligence software. They were seen in footage on social media that seemed to promote the interests of the Chinese Communist Party.
Jack Stubbs, vice president of intelligence at Graphika, told AFP:
This is the first time we've seen a state-aligned operation use AI-generated video footage of a fictitious person to create deceptive political content.- Jack Stubbs, vice president of intelligence at Graphika
In one video that Graphika looked at, a fake male anchor named Alex criticizes the US for not doing more to stop the gun violence that is happening in the country.
In the second, a female anchor talks about how important it is for China and the US to work together as "great powers."
Advances in AI have raised concerns around the world about how the technology could be used to spread false information and harm people. For example, deep fake images can be made out of thin air, and people can be shown saying things they never said.
These newscasters you may have seen online are not real people
Last year, Meta, which owns Facebook, said it took down a deep fake video of Ukrainian President Volodymyr Zelenskyy telling people to give up their weapons and surrender to Russia.
China didn't say anything right away about Graphika's report, which came out just a few weeks after Beijing passed strict rules to control deep fakes.
Last month, China put in place new rules that require businesses that offer deep fake services to find out the real names of their users.
They also want deep fake content to have the right tags so there isn't "any confusion."
Deepfakes are a "danger to national security and social stability," the Chinese government has said.
In its report, Graphika said that the two Wolf News anchors were almost certainly made with technology from Synthesia, an AI start-up in London.
Synthesia's website advertises software for making deep fake avatars "based on video footage of real actors." The company didn't answer AFP's request for comment right away.
Graphika said it found the deep fakes on Twitter, Facebook, and YouTube while looking into pro-China "spamoulage" operations to spread false information.
Mr. Stubbs said:
Spamouflage is a pro-Chinese influence operation that predominantly amplifies low-quality political spam videos.- Mr Stubbs
Despite using some sophisticated technology, these latest videos are much the same. This shows the limitations of using deep fakes in influence operations - they are just one tool in an increasingly advanced toolbox.- Mr Stubbs
Graphika said that the videos they had found were of low quality and "spammy," and none of them had more than 300 views.
It said that this showed "this actor's long-standing challenges in producing convincing political content that generates authentic online engagement."