Meta AI Chief: ‘AI Will Amplify Human Intelligence, Not Take Over Humanity’

At Meta’s Build with AI Summit, Chief AI Scientist Yann LeCun emphasized that AI is intended to enhance human intelligence, not replace it, describing AI assistants as empowering tools rather than threats. He highlighted India’s significant role in advancing AI globally and introduced Meta’s Llama-powered assistant and AI Studio, enabling users to create personalized avatars. Although Meta aims for human-level intelligence in future AI, current large language models remain far from this goal.

At Meta’s first *Build with AI Summit* in Bengaluru, Yann LeCun, Meta’s Chief AI Scientist, clarified that artificial intelligence is designed to enhance human intelligence, not replace it. “AI is not going to dominate or take over humanity; it’s going to amplify human intelligence,” LeCun explained, describing AI assistants as similar to having a highly knowledgeable team available to users. As these systems advance, they may even surpass human intelligence in some areas, but LeCun believes this should empower rather than threaten us.

LeCun emphasized India’s important role in AI’s global evolution, noting that India will contribute not only to local AI solutions but also to international advancements. Meta’s open-source Llama AI model, with over 500 million active monthly users, is projected to be the world’s most widely used AI assistant by the end of 2024, with India leading as its largest user base. Available on platforms like WhatsApp, Instagram, and Facebook, Llama offers support in both English and Hindi, making it accessible across the country.

Additionally, LeCun spoke about the significance of open-source AI, predicting it will become foundational as AI tools integrate globally. To support this, Meta launched *AI Studio* in India—the first market outside the U.S.—allowing users to create Llama-powered AI avatars tailored to their interests, which can act as personalized digital companions.

While Meta’s generative AI capabilities are advancing rapidly, LeCun acknowledged that even large language models (LLMs) are still limited in intelligence. Current models, he noted, fall short of a four-year-old’s cognitive abilities, indicating that human-level AI remains a distant goal. However, Meta is working on innovative systems like Advanced Machine Intelligence (AMI), designed to overcome these limitations, with the ultimate objective of creating AI that approaches human understanding.