Future social media moderation could be AI-driven
As social media companies face an increased risk of regulation, some turn to new methods to manage their?platforms. Community-driven or AI-based rankings could become a common way for platforms to regulate themselves. Facebook has come under fire in recent years for its role in spreading fake news and election interference. As a result, the company has improved transparency and accountability on its platform. One such effort is the introduction of community-driven or AI-based rankings. Community-managed rankings are a system where users can report and vote on content they deem inappropriate. The ranking is then used to determine content placement on the site. YouTube, Reddit, and Wikipedia have already implemented these types of systems.
The most important law for online speech
The spread of misinformation is skyrocketing, and many are anxious to remove or reform?Section 230. This law provides companies with immunity for user content uploaded on their platforms; however, there may be some drawbacks if we do not consider its use, especially when child sexual exploitation could come into play. The proposed Senate Bill advancing earlier this year would create further carve-outs limiting protection against liability based on community-driven ranking systems like Twitter’s recent efforts.
领英推荐
AI-generated assessments
Twitter?is expanding its “birdwatch” program, which displays user-generated context and ratings alongside tweets to arrest the spread of misinformation. In a pilot study conducted over recent?war-related developments?in 2019, these AI-generated assessments were found to reduce posts by 20 - 40%. With billions of users on social media platforms such as Twitter, who lack sufficient resources for good moderation practices, it becomes difficult to identify and act against harmful actions online via impersonations or manipulation schemes. The AI engines are still in development and cannot handle all aspects of social media, but are rapidly improving.
The future may hold great promise when it comes down to leveraging artificial intelligence. Yet, there is a limit to how much they can do for humans since some tasks require human intervention, such as?diagnosing mental health?from one post on Facebook or Instagram. But given the ever-increasing number of daily social media posts, AI is sure to be increasingly leveraged to keep up with the volume of posts.