SAN FRANCISCO — Facebook on Wednesday said it had taken action against malicious actors in Russia, Iran and Myanmar that had deployed fake accounts and other efforts to manipulate social media users, illustrating anew its vast, global challenge to police the platform for an ever-widening array of disinformation.

A small network of accounts with ties in Iran targeted U.S. users, particularly on religious and geopolitical issues, Facebook said in a blog post, while an operation with links to the Russian military focused its efforts on Ukraine and other neighboring countries. The Kremlin-aligned campaign sometimes relied on private messages to steer pro-Russian narratives to influential social media users, a development that worried researchers, who warned the tactics “could be turned towards US targets.”

Facebook said it removed all of the suspect accounts, pages and groups, pointing to its policy prohibiting government interference and coordinated inauthentic behavior. Even though the disinformation campaigns largely occurred abroad and operated on a small scale, the revelation that Russia in particular continues to try to weaponize the social media site is likely to unnerve U.S. leaders. Many in Washington fear the Kremlin could take aim at the 2020 presidential election, much as its online army did four years ago.

“We are making progress rooting out this abuse, but as we’ve said before, it’s an ongoing challenge,” said Nathaniel Gleicher, the company's head of security policy. “We’re committed to continually improving to stay ahead. That means building better technology, hiring more people and working closer with law enforcement, security experts and other companies."

With Iran, Facebook said roughly a dozen accounts — which it did not directly tie to the nation’s government — shared content on U.S. elections and U.S.-Iran relations, sometimes attempting to contact public figures in the process. They had small followings, and Facebook associated the operation to a network of hundreds of accounts with vast reach that it disabled in January 2019. That effort sought to amplify the reach of content produced by Iranian state media. On Wednesday, Twitter also announced it took down a “small number of accounts” with links to Iran.

The takedowns arrived the same week that digital researchers laid bare the full extent of Iran’s efforts to shape political conversations online. So-called sock puppets of the country and its government had more than 2,200 assets on Facebook, including accounts and pages, that directly affected 6 million users globally, according to data compiled by the Atlantic Council. On Twitter, 8,000 accounts produced roughly 8 million messages, their report found. The analysis did not include the latest instances of inauthentic activity on the two sites.

Unlike Russia, which has sought to promote social discord, Iran’s online barrage has aimed to advance “a distorted truth,” wrote lead researchers Emerson T. Brooking and Suzanne Kianpour. The campaign “exaggerates Iran’s moral authority while minimizing Iran’s repression of its citizens and the steep human costs” of its foreign policy, they said. Such efforts have typically peaked when the United States and Iran have been at odds — such as when Washington launched the military strike that killed Maj. Gen. Qasem Soleimani, one of Tehran’s military leaders.

Still, researchers warned that Iran, like Russia, “may attempt direct electoral interference in 2020 and beyond.” Though they said there is no evidence suggesting this has happened, they warned it does not “preclude future such campaigns based on Iranian interest in achieving rapprochement with the United States.”

Facebook also removed about 80 accounts that primarily targeted Ukraine and that Facebook said had “links to Russian military intelligence services.” The operation generally relied on “realistic-looking accounts” to extend the reach of pro-Russian narratives on outside blogs, according to an analysis of the content by Graphika, a research firm that had access to Facebook data. Sometimes, the Russian accounts directly pitched their blog posts to local news sites, “tapping into these outlets’ audiences,” researchers wrote.

Facebook’s new discovery about malicious Russian activity arrives nearly four months after it found the first signs of Kremlin-aligned interference in the 2020 race. The tech giant discovered accounts appearing to be run by locals weighing in on political issues in swing states, attacking former vice president Joe Biden, pitting Democrats against one another and praising President Trump.

Facebook also announced Wednesday it had disabled a novel kind of disinformation operation: an effort by telecom providers in Myanmar and Vietnam to disparage competitors online.

The company Mytel was behind Facebook pages that took aim “at the company’s own competitors as an indirect means of boosting its own brand and profit,” according to the Atlantic Council’s Digital Forensic Research Lab. That included a coordinated campaign to get Facebook users to dial a special code that had the effect of disabling some outbound calls. A day before it surfaced, though, Mytel’s own Facebook page “conveniently” offered viewers the fix for the problem.