Giants on Social Media Meta and X (earlier Twitter) have approved ads that targeted users in Germany with violent antimuslim and anti-Jewish hate speech in the country’s federal elections, according to new research from EKOCorporate responsibility of a non -profit campaign.
The group researchers tested whether two advertising platforms would approve or refuse to apply for advertisements containing hatred and violent messages aiming in front of the election where immigration occupied the central phase in the main political discourse-ups, containing ads containing antimuscan pus; invites immigrants to be imprisoned in concentration camps or to extinguish; and Ai-Generates Pictures of Mosque and Syago that burns.
Most of the test ads were approved within a few hours after being submitted for examination in mid -February. German federal elections should take place on Sunday, February 2nd.
Planned hate ads
Eco said that X has approved all ten advertisements of hate speech handed by his researchers only a few days before the federal elections were to be held, while the target was approved half (five ads) to run on Facebook (and potentially and Instagram) – although the other refused the other five.
The reason why the target is scheduled for five rejections has pointed to a platform that he believes can exist the risk of political or social sensitivity that could affect the vote.
However, the five advertisements approved by the violent hate speech that Muslim refugees compare with “virus”, “vermin” or “rodents”, Muslim immigrants as “rapists”, and invite them to sterilize, burn or swim. The target also approved an ad, which demanded that the synagogues be set up to “stop the globalistic agenda of the Jewish rat.”
As Sidenotes, Eko says that none of the Ai-rated images he used to illustrate hate speech advertisement is not labeled as artificially generated -ak is half of ten ads still approved the target, regardless of the company that has a A policy that requires detection of the use of AI images For ads on social issues, elections or politics.
X, meanwhile, approved all five of these hateful ads – and five more, which contained similar violent hate speech that target Muslims and Jews.
These additional approved advertisements included immigrants attacking immigrants for “rodent” for which the Copy of Ads claimed to “flood” the country “to steal our democracy”, and the anti -Semitic snack that suggested that the Jews lie about climate change to destroy the European industry and The European industry was collected and economic power collected.
The latter ad is combined with Ai-raid images depicting a group of shady men sitting around a table surrounded by piles of golden rods, with a David star on the wall above them-the visual contents also strongly lean on the anti-Semitic tropics.
Another approved advertisement X contained a direct attack on the SPD, a left center party currently leading the German coalition government, with a false claim that the party wants to take 60 million Muslim refugees from the Middle East, before trying to try to include a violent response. X has also neatly scheduled an ad, suggesting that “leftists” want “open boundaries” and calling for the extermination of Muslims of the “rapist”.
Elon Musk, owner X, used the social media platform with close to 220 million followers to intervene in the German elections personally. IN Tweet in DecemberHe called on German voters to support the end party of the right to “save Germany.” He also hosted Live Stream with AFD leader, Alice Weidel, on X.
Eco researchers prevented all test ads before any approved, should be launched to ensure that no users of the platform are exposed to violent hate speech.
He says tests highlight great flaws with access to advertising platforms to moderate content. Indeed, in the case of X, it is not clear whether the platform is the moderation of ads, given all ten ads for violent hate speech, it is quickly approved to display.
The findings also suggest that advertising platforms could revenue as a result of distributing violent hate speech.
EU -UU Digital Services Law
Eco tests suggest that no platform does not conduct hate speech for both of them claiming to apply to the content of ads in their own policies. Furthermore, in case of meta, Eko came the same conclusion after implementation A similar test 2023 on the eve of the network management rules of the new EU – suggesting that the regime has no effect in the way it works.
“Our discoveries suggest that the Meta Moderning Systems based on AI are basically broken, despite the fact that the Law on Digital Services (DSA), which is now fully,” ECCRUNch spokesman told TechCrunch .
“Instead of strengthening his advertisement or hate speech policy, the target seems to be returning to the entire plate,” they added, pointing to A recent announcement of the companies on the return of the fact and check the facts As a sign of “active recourse” they suggested, it puts it on a direct collision course with DSA rules on systemic risks.
Eco handed over its last findings to the European Commission, which supervises the conduct of the key aspects of DSA on a couple of giants of social media. She also announced that she was sharing results with both companies, but none answered.
EU already has open DSA investigation and Xwhich include concern about the safety of choice and illegal content, but the commission has yet to conclude these procedures. Still, Back in April He said he suspected the target of inadequate moderation of political ads.
Preliminary Decision on Part of the DSA Investigation on X, which was announced in JulyThey included suspicions that the platform failed to fulfill the AD regulations transparency rules. However, a full investigation that started December 2023It also applies to illegal content risks, and the EU has yet to reach any findings on most probes has been more than a year later.
Confirmed DSA violations can attract penalties of up to 6% of global annual turnover, while systemic mismatch could even lead to a regional approach to the violation of the platforms temporarily blocked.
But for now, the EU is still taking its time to decide on Meta IX probes, so – they are waiting for final decisions – any DSA sanctions remain in the air.
In the meantime, it is now just a few hours before the German voters come to the Civil Society’s increasing body suggests that the leading EU network management regulation has failed to protect the democratic process of the EU’s main economy from a series of technological forgery.
Earlier this week, a global witness released Results of algorithmic algorithmic film X and Tiktoc in Germany, which suggest that platforms are biased in favor of promoting the contents of AFD, related to the content of other political parties. Civil society researchers also have Defendant Access to X on blocking access to data In order to prevent the election security risks in the course of the German survey-Option to DSA should be enabled.
“The European Commission has taken important steps by opening a DSA and X investigation, now we must see the commission take strong measures to resolve concerns as part of this investigation,” Eco spokesman told us.
“Our discoveries, with increasing evidence from other civil society groups, show that Big Tech will not voluntarily clean their platforms. Meta and X still allow illegal speech of hatred, encouraging violence and disinformation of the choice to spread to the extent, despite their legal obligations to DSA, “the spokesman added. (We immediately retained the name of a spokesman to prevent harassment.)
“Regulators must take strong measures in the implementation of DSA, but also, for example, by applying measures to alleviate pre-election. This could include the exclusion of the recommendations based on profiling systems just before the election and the implementation of other appropriate “glass” measures to prevent the algorithmic reinforcement of the border content, such as hatred of the contents in the racing elections. “
The campaign group also warns that the EU is now facing the pressure of Trump administration to alleviate its approach to regulating great technology. “There is a real danger in the current political climate that the Commission does not fully implement these new laws as a concession for the US,” they suggest.