In the age of algorithms and social media, countering hate speech and disinformation has become a daunting challenge for governments and civil society, especially with the use of AI tools for dissemination by non-state actors.
Multiple surveys and study reports have pointed towards the dangerous trend of stereotypical bias amplification within AI systems, amplification of radical content through algorithms and sophisticated, pattern-wise use of AI tools for misinformation on popular social media platforms.
In that regard, a high-level event marking the Fifth International Day to Counter Hate Speech was held on Thursday, that saw senior United Nations officials and member states reiterate their commitment to countering hate speech and promoting tolerance.
“Hate speech is a growing threat around the world, driven by fear, misinformation, inequality and decreasing trust in institutions… said Miguel Ángel Moratinos, UN Under-Secretary-General, High Representative for the United Nations Alliance of Civilizations (UNAOC), and Special Envoy to Combat Islamophobia, addressing the assembly.
“digital platforms and emerging technologies such as artificial intelligence are speeding up the dissemination of harmful narratives and reinforcing existing prejudices,” he added.
Similar sentiments were raised by H.E. Mohamed Abushahab, Ambassador and Permanent Representative, UAE.
“Hate speech deepens polarization, erodes social cohesion, and contributes to conflict. Left unchecked, it can dehumanize entire communities and undermine the trust on which peaceful societies depend. This threat is also evolving, becoming increasingly global and spreading at unprecedented speed. Extremists exploit digital platforms and emerging technologies to amplify hate speech, incitement, misinformation and disinformation, including through AI-generated content,” said H.E. Abushahab.
Highlighting the UN initiatives, Moratinos discussed how there is now a new e-learning module available on antisemitism that will be available to UN employees starting June 22, 2025, through the UN Campus platform.
He also stated that the consultations are currently ongoing regarding a new UN Action Plan that combats Islamophobia; this will help to build up prevention, education and institutional resiliency against anti-Muslim hatred.
AI and Hate Speech: What Does Studies Say?
Multiple study reports by researchers paint a vivid picture of how artificial intelligence as a technology comes with reinforced bias, often attributed to training, and its use to disseminate misinformation through deepfakes and AI generated slop media content.
A study titled, “Characterizing AI-Generated Misinformation on Social Media” by Chiara Drolsbach and Nicolos Prollochs focuses 91,452 misleading posts (both AI-generated and non-AI made) across 60 languages on X social media platform, flagged through “community notes (CN) system:
The study found out that only 5.06% of them were AI-generated however these posts were disproportionately more likely to go viral despite frequently emerging from smaller accounts on X.
Similarly, another case study points out that automated hate speech detection content affects marginalized communities more.
Another study on “Bias Amplification through GenAI” by researchers from University of Aberdeen and University of Western Australia AI character recommendations significantly amplified people’s baseline stereotyping, and crucially, stereotype-reinforcing influences propagated more readily than stereotype-challenging ones.
Also Read: Breaking News, Broken Answers: NewsGuard Audit Finds AI Chatbots Repeat False Claims
