Germany has gone live with one of the most onerous laws aimed at forcing Facebook Inc., Twitter Inc. and YouTube to police content on their platforms.
The verdict after 10 days in effect? It's complicated.
Since Jan. 1, technology companies face fines of up to EUR50 million ($60 million) if they fail to delete illegal content on their platforms, ranging from slander and libel to neo-Nazi propaganda and calls to violence. The law applies to most social-media networks in Germany.
The banned content was always illegal. What's new is that social networks with more than two million users in Germany now are responsible for cleaning it up themselves.
The new law pushes U.S.-based social-media platforms in Germany one step closer to the level of responsibility that newspapers and media here have long faced -- a level far higher than what the platforms have faced domestically. Under U.S. law, tech platforms aren't liable for user content shared on their services.
Many of the Silicon Valley giants affected by the new rules have already pushed back publicly in Germany. The law has also raised alarm among free-speech watchdogs and legal experts.
"In a democracy, it has to be a state organization that enforces the law, " said Dieter Frey, a lawyer and media expert in Cologne.
Social-media companies typically rely on software and a mix of in-house and third-party content moderators, who sift through posts users have flagged as problematic and delete those that violate local law. In certain cases, the companies' legal teams jump in. The law often has companies working under time pressure to determine whether a post breaches one of 24 paragraphs of the criminal code.
Ahead of the new law, Facebook contracted with providers for 1,200 moderators in Germany, a number that compares with 7,500 moderators world-wide. Just 1.5% of Facebook's 2.07 billion monthly users world-wide are based in Germany.
Facing increased pressure after the U.S. election and terrorist attacks around the globe, tech companies have taken some voluntary steps to monitor the massive amount of content on their platforms. Facebook, for instance, is figuring out how to fully monitor and analyze the more than one million user reports of potentially objectionable content that it says it receives every day.
At YouTube, a unit of Alphabet Inc.'s Google where users watch more than a billion hours of video a day, the company has used both software and humans to screen for content that warrants removal, such as extremist videos. In the U.S., Google has said it instructs human reviewers to mark violent or hateful content as low quality, which will likely move such sites lower in Google search results. Twitter has been using internal technology to flag accounts that promote terrorism.
Following rules can pose practical difficulties, as companies have found in the first 10 days Germany's new law has been in effect.
This month, Twitter temporarily suspended the account of a German satire magazine, Titanic, demanding that the magazine delete a parody tweet mocking a German nationalist lawmaker, Beatrix von Storch.
Tim Wolff, Ttanic's editor in chief, said Twitter notified him of the suspension by email on Tuesday, Jan. 2, and asked him to delete the tweet. Later, other users piled on and flagged four other tweets on Titanic's account, including one that made fun of German police and another about Austria's Chancellor Sebastian Kurz, whom Titanic has called "Baby Hitler" and said should be killed.
Twitter asked Mr. Wolff to delete those four tweets as well. Then, after an internal review that included input from legal professionals on its support team in Germany, Twitter dropped that request and reversed Titanic's suspension. The tweet about Ms. von Storch remains offline.
Mr. Wolff said, "Our suggestion is to let us at Titanic decide what is satire and what isn't."
In another case, on December 22, before the law had taken full effect, Facebook blocked the account of Mike Samuel Delberg, a 28-year-old political representative of Berlin's Jewish community, after he posted a video of an Israeli restaurant owner in Berlin being threatened on the street.
In a false positive, Facebook content monitors thought the video violated the company's community standards. "It went viral," Mr. Delberg said. Then "all of a sudden Facebook deleted the video and blocked me and said that I broke their guidelines." The company has since apologized to Mr. Delberg for deleting the video and he is back online.
"It can't be true that...while raising awareness in public of anti-Semitism, an account gets deleted," Mr. Delberg said.
"We should not be the ones who judge if a post is illegal or not," said Semjon Rens, Facebook's public-policy manager in Germany. "This is the responsibility of courts." He said Facebook is "working hard to put the right processes in place and to comply."
Google didn't respond to requests for comment. Google executives have said the German law is overreaching.
A representative for Twitter declined to comment on the record or to disclose the size of the team it employs to review content on its site but did explain the company's views.
German enforcement officials are still finding their way. Ulf Willuhn, a senior public prosecutor in Cologne, spent the beginning of the year considering whether his office would have to pursue legal action based on a tweet from an anti-immigrant lawmaker -- the real Ms. von Storch, not a parody this time -- who had referred to the local Arabic-speaking community as "group rapists."
After some research, Mr. Willuhn decided his office wasn't responsible. Separately, Twitter had already asked her to delete her tweet.
"At the end of the day, ...it's not as easy as it seems at first glance, " Mr. Willuhn said. "Yeah, it's very very difficult."
Write to Zeke Turner at Zeke.Turner@wsj.com
(END) Dow Jones Newswires
January 10, 2018 12:41 ET (17:41 GMT)