Meta sees limited use of GenAI in Russian disinformation

An EU Commission probe into Facebook and Instagram’s handling of Russian fake news is still pending

Generative artificial intelligence tools used to spread mostly Russian disinformation on Meta’s platforms have not been disruptive, the US tech giant said in its quarterly Adversarial Threat Report published yesterday (15 August). 
The findings suggest that GenAI-powered tactics provide “only incremental productivity and content-generation gains” to the threat actors, but the tools used by Meta to counter the spread of disinformation currently appear to be effective.
This comes as a European Commission probe into Meta’s Facebook and Instagram handling of disinformation under the Digital Services Act (DSA) is still pending. The Commission fears that the networks are vulnerable to Russian misinformation, and are potentially a target for Russian networks.
A company spokesperson said in April that Meta has a “well-established process for identifying and mitigating risks on our platforms” in place. 
In its report, Meta said that the Russian campaign published a large volume of stories trying to mimic authentic articles from across the internet, including mainstream media, on fictitious ‘news’ websites. 
These stories were likely summaries of the originals generated using AI tools to make them appear more unique, the Meta report said. The same campaign then also posted AI-generated news-reader videos on YouTube and ran fictitious journalist personas.
Since the start of Russia’s war in Ukraine in 2022, the fake news campaigns largely focus on undermining Ukraine at home and abroad, though some networks also focused on other countries at Russia’s border like Georgia and Moldova.
The report said that, ahead of the US elections in November, Russia-based operations are expected to promote content supportive of presidential candidates who oppose aid to Ukraine. This could include blaming the US for providing financial help to Ukraine, presenting Ukraine’s government as unreliable, or amplifying voices that express pro-Russia views on the war.
Besides Russia, Meta also took down a number of fake accounts and pages in Iran and China.