Facebook has admitted that it observed attempts to spread propaganda on its site, apparently orchestrated by governments or organised parties.
The firm has seen “false news, disinformation, or networks of fake accounts aimed at manipulating public opinion”, it revealed in a new report.
“Several” such cases during the US presidential election last year required action, it added.
Some of the activity has been of a “wide-scale coordinated” nature.
Fake accounts were created to spread information stolen from email accounts during the 2016 US presidential election, the firm noted, though it said the volume of such activity was “statistically very small”.
But the company added that efforts to tackle “information operations” had led it to remove more than 30,000 fake accounts in France – where a presidential election is currently under way.
In general, Facebook said it faced a new challenge in tackling “subtle and insidious forms of misuse, including attempts to manipulate civic discourse and deceive people”.
Facebook described much of the activity as “false amplification” – which included the mass creation of fake accounts; the coordinated sharing of content and engagement with that content (such as likes); and the distribution of “inflammatory and sometimes racist memes”.
It added, however, that there was not much evidence that automated bots had been set up to do this, but humans appeared to be directly involved.
“We have observed many actions by fake account operators that could only be performed by people with language skills and a basic knowledge of the political situation in the target countries, suggesting a higher level of coordination and forethought,” the report explained.
The apparent objectives of those behind the propaganda efforts included one or more of the following:
- Sowing distrust in political institutions
- Spreading confusion
- Promoting or denigrating a specific cause or issue
Facebook said that it was working on a variety of methods to curb the spread of propaganda on its platform.
These included building new products to help stamp out fake news and creating new systems – some with artificial intelligence capabilities – to help quicken the response to reports of fake accounts or spam.