Need global compact to prevent AI misuse: PM Modi at G20
Prime Minister Narendra Modi said that the world needs a “global compact on AI”, which should include effective “human oversight, safety-by-design, transparency”, and strict prohibitions on the use of AI for “deepfakes, crime, and terrorist activities”.
Addressing G20 leaders during a session on “A Fair and a Just Future for All – Critical Minerals; Decent Work; Artificial Intelligence”on the second day of the summit in Johannesburg, PM Modi also extended an invitation to all G20 countries to participate in the AI Impact summit to be held in India in February 2026.
“Under the India-AI Mission, we are developing accessible high-performance computing so that the benefits of AI reach every district and every language. This will provide both scale and speed to our efforts toward human development,” the Prime Minister said.
“At the same time, we must all ensure that AI is used for the global good and that its misuse is prevented. For this, we need a global compact on AI, built upon certain fundamental principles. These should include effective human oversight, safety-by-design, transparency, and strict prohibitions on the use of AI for deepfakes, crime, and terrorist activities,” Modi said, highlighting the basic principles.
“AI systems that impact human life, security, or public trust must be responsible and auditable. And most importantly, while AI may enhance human capabilities, the ultimate responsibility for decision-making must always remain with human beings,” he said.
In this age of AI, the Prime Minister said, “We must swiftly shift our approach from the ‘jobs of today’ to the ‘capabilities of tomorrow’. Unlocking talent mobility is essential for rapid innovation. We made progress on this issue at the New Delhi G20 Summit. We hope that in the coming years, the G20 will develop a global framework for talent mobility.”