TASI 2026: AI safety, child protection take centre stage at panel discussion

AI technology is progressing at a rapid pace, but the safeguards intended to protect users and society are struggling to keep up, according to a panel discussion at the Trust and Safety India (TASI) Festival in New Delhi. The panel, composed of senior executives from leading tech companies and policy experts, examined a range of AI risks including deepfakes, child safety, and privacy concerns.

They debated whether companies can manage these risks through self-regulation or if stricter government oversight is necessary. The panel also explored issues like red-teaming, industry-wide safety sharing, gaps in India’s regulatory framework, and culturally sensitive AI evaluations. Key takeaways include the need to prioritize AI safety as models become more capable, the challenge of balancing company incentives with competition, and the importance of testing AI before release, especially for vulnerable populations such as children.

The panel highlighted the necessity for AI systems to understand and respect local languages, cultural contexts, and digital skills, while also avoiding restrictions on access to information and freedom of expression.