presentation reviews#

Hate Speech on Social Media

Presented By: Anthony Bowen, John Drew, Evan Hamp, Brett Johnson, Kavel Mason, Xiaoyo Niu, and Chris Robb

Problem: Hate Speech on Social Media

Elements of the problem:

● 1st Amendment ● Defining hate speech ● Existing policies in place ● International policies on speech ● Social media platforms content algorithms

Problem Statement

Hate speech is the use of expression or communication to antagonize a person or group due to ethnicity, race, nationality, sexuality, or other traits of an identity. Hate speech has become common in most online platforms but the biggest issue is that there is a transparent lack of accountability on all sides (Platforms and Users).

Alternatives ● Comment AI

○ Platforms develop algorithms to automatically flag or block posts deemed as hate speech using keywords and patterns of user behavior. ● Account Bans

○ Using data collected using Comment AI and user feedback, platform employees would be able to easily identify and ban accounts for repeated incidences of hate speech.

● Platform Accountability ○ Through intervention of the Congress, repeal Section 230 of the CDA which would return responsibility to companies for the content posted to their

platforms.

● Positivity Feedback ○ Platforms add functionalities that allow users to give more in-depth feedback on posts than the current “like” system, and in turn algorithms would

promote posts with positive feedback.

● Legal Intervention ○ Require users to register their account(s) with their legal ID’s, which would subject users to possible legal repercussions from law enforcement.

Alternatives’ Criteria

● Doesn't make any normal activities more challenging or limit use ● Easy Implementation/ explanation ● Reliability ● Maintains freedom of speech ● Reasonable Consequences

Outcomes

Trade Offs

● Comment AI, Legal Intervention, and Account bans potentially violate user rights

● Using the upvote system and User Comment Preferences does not remove hate speech

● Universal Frequency Limits would limit all of users

● Platform Accountability Violates the legal rights of any platform involved

● Most of these alternatives assume that majority of users would be against the same form of hate speech.

● Positivity Upvotes can amplify positive voices thereby rendering hate speech ineffective

● Legal intervention has shown to be effective in regions that utilize it.

● Universal frequency Limits would force users to post important content.

● Comment Preferences can be used to protect individuals who otherwise would be at the mercy of hate speech.

Decide with even swaps By implementing each criteria with solutions, we were able to determine how reliable those solutions were and how they may affect the outcome of each criteria. We were able to consider the problems faced.

We were also able to determine the level of dominance between each alternative in an exact order:

Preserves rights, Manageable Risk, Reliable,User Friendly, Easy Implementation

Stakeholders

stakeholders have some way to benefit off of those who are the consumer.

Most interest groups tend to fall under the civil society rather than the private sector.

Works Cited

Cash, Meredith. “Sarah Fuller Said Haters Can 'Talk Crap All They Want' about Her Historic Stint with Vanderbilt Football Team.” Insider, Insider, 30 Nov. 2020, www.insider.com/sarah-fuller-vanderbilt-critics-can-talk-crap-all-they-want-2020-11.

“Hate Speech on Social Media: Global Comparisons.” Council on Foreign Relations, Council on Foreign Relations, www.cfr.org/backgrounder/hate-speech-social-media-global-comparisons.

“Hate Speech Regulation on Social Media: An Intractable Contemporary Challenge.” Research Outreach, 17 Mar. 2020, researchoutreach.org/articles/hate-speech-regulation-social-media-intractable-contemporary-challenge/.

“Tay, Microsoft's AI Chatbot, Gets a Crash Course in Racism from Twitter.” The Guardian, Guardian News and Media, 24 Mar. 2016, www.theguardian.com/technology/2016/mar/24/tay-microsofts-ai-chatbot-gets-a-crash-course-in-racism-from-twitter.