Microsoft reveals that AI-generated news anchors are used in disinformation efforts in Taiwan

Microsoft has revealed the emergence of AI-generated news anchors used to spread disinformation. The information is based on recent efforts which are targetted at Taiwan’s election.

The information reveals that the Chinese state-backed groups as main part in this news. These groups are said to be creating fake news videos which feature AI generated anchors who deliver biased or misleading information about political figures.

One example cited involves an AI-generated anchor criticizing Taiwan’s president with unsubstantiated claims related to family. 

In one clip cited by Microsoft, the AI-generated anchor made unsubstantiated claims about the private life of the ultimately successful pro-sovereignty candidate – Lai Ching-te – alleging he had fathered children outside marriage.

Microsoft also links these efforts to a video editing tool, CapCut, developed by the Chinese company ByteDance, which owns TikTok.

Experts consulted by Microsoft talks about the limitations in the current generation of AI-generated anchors. These videos often exhibit unnatural movements, unrealistic voices, and lack details like natural lighting on the figures. I wonder how close they are to the lastest unveiled GPT-4o

Despite these limitations, some social media users have been misled by the videos. However, the good news is that the report finds no major impact on public opinion, as of now at least, as evidenced by the outcome of Taiwan’s election.

The information suggests a potentially greater threat lies in manipulating existing footage of real news anchors. This could involve doctoring videos to make journalists appear to say something untrue. Microsoft also warns of the potential future development of more convincing AI-generated video technology, which, in my humble opinion, is bound to happen.

Another Microsoft-related news for Chinese users: Chinese Windows 11 Insiders get PC Manager.

More here.