Some conservative Twitter Inc. users are accusing the company of unfairly targeting their followers as it steps up efforts to get rid of fake or abusive accounts.

Prominent conservative pundits and activists said today that thousands of their followers had been deleted overnight. Other users said they received messages from Twitter asking them to confirm they were real people before being allowed to keep using the service.

The hashtag #TwitterLockOut was trending in the United States.

“The twitter purge is real,” conservative podcast host Dan Bongino said on Twitter. “Twitter blocked me from twitter ads last night and purged thousands of followers.”

White nationalist Richard Spencer and conservative broadcaster Bill Mitchell both said they had been affected, too.

Twitter said that when it identifies an account that may violate its terms of service, it will ask the account owner to confirm a phone number to verify that a human is behind the account. That’s why some people may be experiencing suspensions or locks, the company said. When an account is locked and being challenged to provide a phone number, it is removed from follower counts until it provides a phone number, Twitter added.

“Twitter’s tools are apolitical and we enforce our rules without political bias,” the company said in a statement. “As part of our ongoing work in safety, we identify suspicious account behaviors that indicate automated activity or violations of our policies around having multiple accounts, or abuse.”

Verifying a phone number also helps San Francisco-based Twitter find violators who are operating multiple accounts for abusive purposes, according to the company’s policies. In 2016, Spencer was temporarily removed for having multiple overlapping accounts.

Twitter is escalating efforts to combat malicious bots — or automated accounts — that artificially inflate follower counts and advertising metrics. Twitter was designed to be friendly to innocuous bot accounts that can help companies quickly spread their message or respond to customer-service complaints. Yet bots can also be used to pose as other people’s identities or to spread misinformation.

Some fake accounts have been traced to Russian-backed agents that the U.S. government says are working to sow political discord. Russian-linked Twitter bots shared Donald Trump’s tweets almost half a million times during the final months of the 2016 presidential election. Researchers say as many as 15 percent of accounts could be fake or spam, a number Twitter says is much lower.

Conservatives have long accused Twitter of targeting them specifically. Even Ajit Pai, the chair of the Federal Communications Commission, has said the service discriminates against conservatives. On the flip side, progressive users say Twitter doesn’t do enough to stop harassment against women and people of color. Some argue that President Trump, Twitter’s most influential user, should be banned for bullying opponents.