Photograph by Dominic Lipinski — PA Wire/AP

Online abuse has been a problem ever since the Internet was created. But over the past few years, it seems to have escalated—despite the efforts of platforms like Twitter and Facebook to try and control it.

And some experts believe it could get worse before it gets better.

The question asked by the researchers: “In the next decade, will public discourse online become more or less shaped by bad actors, harassment, trolls, and an overall tone of griping, distrust, and disgust?” Over 40% replied they don’t expect this situation to change much, and another 39% said they could see it actually becoming more of a problem rather than less.

“Trolling will continue, while social platforms, security experts, ethicists, and others will wrangle over the best ways to balance security and privacy, freedom of speech, and user protections,” Susan Etlinger, a technology analyst at the Altimeter Group, told Pew researchers.

Get Data Sheet, Fortune’s technology newsletter.

Although certain online “safe spaces” may be developed that will be free of trolls and harassment, some of the experts surveyed said that these will be little more than Potemkin villages—that is, attractive facades that hide the true nature of the social web.

In some cases, the research report warns an attempt to control abuse and harassment could actually result in an infringement of personal freedoms, including freedom of speech, and could lead to the web becoming less open and more polarized.

“One of the biggest challenges will be finding an appropriate balance between protecting anonymity and enforcing consequences for the abusive behavior that has been allowed to characterize online discussions for far too long,” said Bailey Poland, author of the recent book Haters: Harassment, Abuse, and Violence Online.

This is a problem that Twitter in particular has struggled with for much of its life. The company’s senior executives often stressed that the service was “the free-speech wing of the free-speech party,” and that users should be free to say whatever they wished anonymously, and some believe that hampered its ability to address abuse on the platform.

Some of those who responded said that they expect better reputation-management systems and moderation tools may help to solve the problem, but others said they fear that these kinds of tools will remove anonymity and make surveillance (including government monitoring) easier.

There was a certain fatalism underlying many of the responses that expect the situation to remain unchanged, the Pew Center found.

“Social media will continue to generate increasingly contentious, angry, mob-like behavior,” said Paul Edwards, a professor of information and history at the University of Michigan. “The phenomenon that underlies this behavior has been consistently observed since the early days of email, so there is no reason to think that some new technique or technology will change that.”

An anonymous respondent told the Pew researchers that “human nature has not much changed over the past 2,000 years; I don’t expect much change over the next 10.”

In a recent essay, Microsoft sociologist danah boyd (who spells her name using only lowercase letters) said much the same thing about the problem of “fake news” or false information online. Although many people wish that Facebook and Google could fix the problem, she explained, this is impossible because it is a social and cultural problem.

“No amount of ‘fixing’ Facebook or Google will address the underlying factors shaping the culture and information wars in which America is currently enmeshed,” boyd says.

A number of respondents to the Pew study also noted that there is an economic incentive for social platforms and websites to encourage polarizing content, including fake news, because it drives engagement and thereby boosts revenue, which is dependent on advertising. “Technology companies have little incentive to rein in uncivil discourse,” the report says.

Fake news and similar problems are also being fueled by the fact that governments and other political forces have found they can manipulate people’s behavior, argues Laurent Schüpbach, a neuropsychologist at University Hospital in Zurich.

“The reason it will probably get worse is that companies and governments are starting to realise that they can influence people’s opinions that way,” he told the Pew researchers. “And these entities sure know how to circumvent any protection in place. Russian troll armies are a good example of something that will become more and more common in the future.”

Mathew Ingram

Believe It or Not: Online Trolling and Abuse Could Get Worse