Misinformation and disinformation date back thousands of years, but in the Information Age, both types of falsehoods proliferate. Why? Every day, billions of people and millions of organizations generate more and more data on an expanding universe of connected devices. By one estimate, we generate 2.5 quintillion bytes of data each day. As the information ecosystem expands, mis/disinformation increases.
What is misinformation & disinformation?
Both misinformation and disinformation fall under a broader category of inaccurate or false information. But the primary difference between the two is a matter of intent.
Disinformation is false information that is intended to mislead. Sometimes we call disinformation propaganda, but it’s important to recognize that a range of bad actors, from criminal gangs to terrorists to nation-states, run disinformation campaigns.
Misinformation is spread without a specific agenda. The information could be 100% false, or it could be a mix of facts and falsehoods. But unlike with a targeted disinformation campaign, misinformation spreads in media environments where there’s a dearth of accurate information and multiple diverging assessments as to what the truth is.
The harm caused by mis/disinformation
Mis/disinformation impacts every aspect of society. For example, mis/disinformation can undermine confidence in public health measures, elections, or a particular commercial product. Mis/disinformation can also amplify conspiracy theories that lead to increases in hate speech and violent crime. Left unchecked, the widespread proliferation of mis/disinformation corrupts a free society by attacking the very nature of truth itself.
Social networks are the breeding ground for mis/disinformation
While mis/disinformation can spread anywhere, social media networks are particularly fertile areas for falsehoods. As MIT researchers found when they analyzed 126,000 stories between 2006 and 2016, false stories spread six times faster than true stories on social networks. Unfortunately, as social media use grows, opportunities for mis/disinformation also grow.
What impact do bots have on mis/disinformation?
Bots are a relatively cheap, easy-to-use tool for coordinating and amplifying information campaigns. When those same bots are used to spread mis/disinformation they can cause havoc.
How big is the bot threat? One estimate from TwitterAudit found that only 40 to 60% of Twitter accounts represent real people, meaning that the remaining accounts may be bots. But bots aren’t unique to Twitter. All social networks are vulnerable to bots spreading mis/disinformation.
Scanning multiple data sources to determine the credibility of information
To determine the credibility of information, analysts must look at multiple data sources. To do so at scale, analysts rely on AI tools to focus their efforts. Here are four key data points analysts typically consider when assessing credibility within different data sources.
Are the relationships within the information network unnaturally dense?
Bot networks almost always sit at the periphery of the real conversation. By analyzing secondary and tertiary connections, analysts can determine whether the conversation is organic or a “master” account is pulling the strings.
Are post-generating algorithms at work?
Typical human interactions involve a mix of original content, reposts, and engaging with or replying to conversation streams. In contrast, bots rely heavily on reposts, but rarely share original content or engage in conversation.
Is the posting schedule highly uniform?
Humans post for a variety of reasons, based on our moods, desires, and activities. As a result, our posts are somewhat random. Bots, on the other hand, post on a highly uniform schedule. Analyzing the posting schedule reveals patterns that are inconsistent with human behavior.
Is the information positioned to influence a specific audience?
Often, the target of a bot is identifiable because bot networks are tools designed to achieve specific information goals. By looking at data sources, then looking at the information from the perspective of its intended audience, analysts can determine the credibility of information with a high degree of confidence.