Main content

Amber Heard vs the Internet: An Organised Smear Campaign?

Alexi Mostrous uncovers evidence that Amber Heard was targeted by a coordinated online trolling campaign, with fake pro-Depp accounts flooding social media ahead of the defamation trial.

On the surface, they looked like Johnny Depp superfans.

Social media profiles overflowing with praise for the Hollywood actor – and anger against his ex-wife, Amber Heard.

In reality, these accounts weren’t genuine at all. Evidence we collected in our podcast Who Trolled Amber? – now available on 蜜芽传媒 Sounds – suggests they were controlled by human trolls with links to the Saudi government.

In 2022, Depp won a defamation case against Heard after a U.S. jury found she had made up allegations of domestic abuse against him. Depp’s victory was a huge comeback for the Pirates of the Caribbean star. Two years before, a UK court had found that Depp had abused Heard on 12 occasions.

Before the U.S. trial, thousands of pro-Depp accounts sprang up on social media. Heard’s team suspected someone had hired automated accounts to attack her, but they could never prove it.

Were they right after all?

Listen to Who Trolled Amber? on 蜜芽传媒 Sounds

Image by Getty Images

In the six-part podcast, we find evidence to suggest that Heard was subjected to an organised trolling campaign.

We obtained a cache of almost one million tweets posted about the actor in the run-up to the trial. One data expert who we commissioned to look at the cache told us that more than 50 percent of these tweets were inauthentic.

According to the expert, that means they either came from bots, which are automated accounts, or paid for “trolls” - real people hired to slander someone online.

In one case, 100 accounts sent 1,000 identical messages at once to companies working with Heard saying: “This brand supports domestic violence against men.” We also found bot-networks tweeting pro-Depp content in Thailand and Spain as well as fake pro-Depp accounts that used AI-generated profile pictures to appear authentic.

Alexi Mostrous, Investigations Editor at Tortoise

Then we found a small network of pro-Depp accounts which were particularly suspicious. When we found them, the accounts were tweeting about the actor in English. But using the Wayback Machine, which saves deleted webpages, we found the same profiles had posted hundreds of now-deleted tweets in Arabic. None of these posts mentioned Depp. Instead, most praised the Saudi regime.

Depp has visited Saudi on several occasions and is a personal friend of Mohammed Bin Salman, the country’s ruler. Saudi has also financed the actor’s two most recent films - Jeanne du Barry and Modi. And Saudi Arabia has a long history of using bots to manipulate social discourse.

Depp and his team have denied wrongdoing. There is no evidence that any of them were involved in setting up any fake accounts targeting Heard. And the number of Saudi accounts we found was relatively small.

But the implications are significant. While we don’t know if any of the social media activity affected the US Depp vs Heard trial, we do know that the jury wasn’t sequestered, meaning they could technically read what was being said about the case on the internet.

More widely, what we discovered about online manipulation goes beyond one celebrity case. It shows us how easily online disinformation can be created and published on an industrial scale, and how difficult it is to detect.

Since we made the podcast, AI-based tools have made it almost impossible to distinguish between what’s real and fake online.

What happened to Amber Heard is, in a way, only the start.

Listen to Who Trolled Amber? now on 蜜芽传媒 Sounds.

More from Tortoise Investigates