AI Program Developed To Tackle Hate Speech Menace in India

by

As a seasoned analyst with extensive experience in digital media and technology, I am heartened to see the proactive steps being taken by experts to combat the growing menace of hate speech and deepfakes on social media platforms. The ongoing efforts at Dhirubhai Ambani Institute of Information and Communication Technology (DAIICT) are particularly noteworthy, given India’s significant presence as a consumer of social media giants.

In order to combat the increasing issue of hate speech dissemination and deepfakes on social networks, particularly concerning India, experts are currently developing an Artificial Intelligence (AI) system designed to identify sophisticated forms of online deception employed by individuals spreading hate speech.

As an analyst, I recently participated in the “Forum for Information Retrieval Evaluation” (FIRE) held at the Dhirubhai Ambani Institute of Information and Communication Technology (DAIICT) in Gandhinagar, Gujarat. Over a span of four days, we delved into the latest trends surrounding large language models (LLM) within Machine Learning (ML), sharing insights and fostering discussions on this rapidly evolving field.

At the event, you’ll find Professor Thomas Mandl from University of Hildesheim. He’s currently collaborating with an Indian organization on a project that identifies patterns of hate speech. According to Professor Thomas, it has become challenging to monitor hate speech text using algorithms over time because the perpetrators often employ mixed languages, use ‘0’ instead of ‘O’, and other coded terms.

As per Professor Thomas’s statements, the AI system under development is designed to identify sophisticated forms of disguised hate speech from potential offenders.

According to Professor Thomas, it’s essential that we regularly revise the system and educate artificial intelligence on recurring patterns, vocabulary, popular themes, trending subjects, and unique demographic sectors. Our efforts encompass not only the English language but also Indian languages as well.

As technology progresses and social media expands, there’s been a significant increase in instances of hate speech in India. This issue has even caught the attention of the Supreme Court. Given that India is among the world’s biggest users of major social media platforms, the potential for hate speech to spread on these platforms and its real-world consequences is particularly severe.

During the gathering, Professor Prasenjit Majumder from DAIICT explained that in recent years, there’s been growing understanding of AI and ML capabilities and how they should be prepared for future applications. One significant endeavor they are engaged in is real-time translation of parliament debates in various Indian languages like Gujarati, Telugu, Tamil, Bengali, Malayalam, and others. According to him, this initiative could revolutionize accessibility, making parliamentary discussions more inclusive for a broader public.

According to Article 19(1)(a) of the Indian Constitution, all citizens are granted the fundamental right to free speech and expression. However, when this freedom is abused, the Supreme Court of India has directed the Law Commission to examine instances where speech exceeds the boundaries of judicial intervention. Notably, many occurrences, such as those happening within religious contexts and online platforms, have singled out minority communities, which has prompted the Supreme Court’s focus on this issue. As a response, they are considering legislation aimed at controlling hate speech in India.

Read More

2024-12-16 16:55