Bloomberg: Hamas attack on Israel was a test for Musk's X - and it failed

Posts about the attack on Israel led to confusion, disinformation and conflict on Elon Musk's Twitter, showing that new ownership and policy changes turned the platform into an unreliable resource during the crisis, researchers say.

Source. Bloomberg reports.

Hours after armed Hamas militants from the Gaza Strip stormed into Israel in the largest attack on the country in decades, unverified photos and videos of air strikes, the destruction of buildings and homes, and other posts depicting military violence – in Israel and Gaza – flooded the social network.

Many posts promoted old images of armed conflicts, passing them off as new, and were distributed by anonymous accounts with blue check marks – a signal that they had purchased verification at the "premium" price of an X subscription.

Other accounts passed off video game footage as military photos. Far-right pundits spread false statements about the conflict that went viral, a common tactic to increase user engagement, the publication writes.

Bloomberg recalls that after Musk's purchase, ex-Twitter weakened its moderation and content policy, the consequences of which became apparent during the current geopolitical crisis.

Thus, a significant number of fakes were promoted by anonymous accounts with a blue check mark. In the "old" Twitter, this sign served as a quality benchmark, while in the new one it has become only a sign of paid verification for $8 per month.

At the same time, fact-checking on the platform is increasingly being handed over to volunteers through the Community Notes service.

The sharing program also incentivizes users to make as many posts as possible without sufficient verification by advertising revenue, which X introduced in July for accounts with a large reach. "I think a lot of them looked at the performance of posts about Russia/Ukraine in 2022 and wanted to repeat it," says Emerson Brookings, an Internet researcher at the Atlantic Council.

"It's about his [Musk's] behavior on the platform – he thinks conspiracy theorizing is okay, and he spreads it himself, and these things come from the top down," says Kayla Gogarty, research director at Media Matters. And the platform failed it."

The European Union is launching an investigation into the social network for misinformation about the recently passed Digital Services Act. The company potentially faces fines of up to 5% of daily turnover for each violation.

"Our policy is open source and transparency, which I know the EU supports. Please list the violations in X to which you refer so that they can be seen by users. Merci beaucoup," Musk replied to European Commissioner Thierry Breton.

Fakes about Israel's war with Hamas were plentiful on other platforms, but on X they have taken on a new quality, says conspiracy theorist Mike Rothschild.

"It's become almost impossible to distinguish between fact and rumor, trolling or conspiracy theories," he says. "Musk's changes have made X not just useless in a crisis, but much worse.

Bloomberg cites several examples of malicious disinformation breaking through the weakened filters. The first is a video allegedly showing Palestinians killing Jewish settlers in their own homes. It was distributed by Malaysian streamer Yang Miles Chong, who personally communicates with Musk.

Before it became clear that Israeli law enforcement officers were in the footage, the video was viewed 12.7 million times. X did not remove it, but only marked it with a community note.

Fakes about the US allocating $8 billion in military aid to Israel and the Taliban's desire to enter the conflict received coverage in the hundreds of thousands and millions.

A post about weapons allegedly received by Hamas from Ukraine was viewed 7 million times, and a video of an Israeli helicopter being hit by an MANPADS taken from the Arma3 game was viewed almost as many times.

David Frum, a former speechwriter for President George W. Bush, wrote in X that 20 minutes in a Twitter feed used to have a clearly higher information value than watching the news on TV.

"Now, there is still useful information out there, but finding it, selecting it, and distinguishing it from fake news is much harder than it was a year ago," Frum wrote.

X denied that in the early days of the conflict, she flagged and deleted tens of thousands of posts with disinformation. This was done by a separate task force led by CEO Linda Yaccarino, who even canceled a speech at The Wall Street Journal's tech conference.

However, the proportion of moderated posts on X was significantly lower than on Facebook, even when adjusted for the size of the audience – 8900 vs. 415 thousand per day, the WSJ notes. In 75% of cases, the post was simply marked with an NSFW (restricted viewing) plate without being deleted.

Background. As a reminder, according to the expert, Kremlin bots made every effort to spread the Musk and Zelensky meme.