As the AI arms race accelerates, Meta AI’s chief AI scientist, Yann LeCun, challenges the popular opinion that AI will surpass human intelligence in the near future. In an interview with CNBC on the 10th anniversary of Meta’s AI lab, LeCun expressed skepticism about Artificial General Intelligence (AGI) and questioned the current capabilities of quantum computing.
By ‘not any time soon’, I mean ‘clearly not in the next 5 years’, contrary to a number of folks in the AI industry. Yes, I’m skeptical of quantum computing, particularly when it comes to its application to AI, clarified LeCun via his LinkedIn.
LeCun’s views sharply contrast with Nvidia CEO’s prediction that AI will match human intelligence within five years. A pioneer in deep learning, LeCun believes that current AI systems are decades away from achieving a level of sentience or common sense. He argues that the industry is more likely to develop cat-level or dog-level AI before reaching human-level intelligence.
LeCun also has reservations about the effectiveness of current AI training methods that heavily rely on text data. He points out that text is a very poor source of information for training AI due to its lack of richness and complexity compared to real-world interactions. In response, Meta is exploring multimodal AI systems that integrate audio, image, and video data, moving beyond language models like ChatGPT.
LeCun’s opinion further contrasts with Elon Musk’s concerns about AI posing an existential threat to civilization. LeCun dismissed Musk’s fears as completely ridiculous, emphasizing that AI becoming smarter than humans doesn’t necessarily mean it will seek to control or harm humanity. Meta AI, under LeCun’s guidance, is focused on developing practical applications of AI that can make a real difference.
Projects like Audiobox and Seamless reflect Meta’s ambition to push AI into new territories such as audio generation and language translation, with the aim of enhancing daily lives. Unlike Google’s DeepMind, which keeps its AI developments under wraps, LeCun intends to democratize AI development with projects like the LLaMA model.
By embracing a grounded approach to the future of AI, Meta AI aims to dispel concerns of an imminent superintelligence takeover and promote the beneficial applications of AI. While the industry continues to debate the timeline for AGI and grapple with the potential of quantum computing, Meta remains committed to advancing the field with practical and impactful AI solutions.
