Sign up to save tools and stay up to date with the latest in AI
bg
bg
1

Meta AI Head: ChatGPT Will Never Reach Human Intelligence

May 22, 2024 - pymnts.com
Meta's chief AI scientist, Yann LeCun, has expressed skepticism about the potential of large language models (LLMs) like ChatGPT to achieve human-level intelligence. LeCun argues that these models lack understanding of the physical world, persistent memory, and the ability to reason or plan. He is instead focusing on developing a new generation of AI systems that could achieve human-level intelligence, a process that could take a decade. This approach is seen as a risky gamble, as investors are looking for quick returns on their AI investments.

Meanwhile, other companies are pushing ahead with advanced LLMs in the hope of creating artificial general intelligence (AGI), machines that surpass human cognition. AI firm Scale recently raised $1 billion towards this goal, and French startup "H" raised $220 million. However, some experts, like Akli Adjaoute, author of "Inside AI", question the ability of AI to think like humans. Adjaoute sees AI as an effective tool, but emphasizes its limitations in replicating human understanding.

Key takeaways:

  • Meta’s chief AI scientist, Yann LeCun, believes that large language models (LLMs) like ChatGPT will never reach human intelligence due to their limited grasp on logic and inability to understand the physical world or reason.
  • LeCun is working on a new cohort of AI systems aiming for human-level intelligence, a process that could take up to 10 years and is seen as a risky gamble by some investors.
  • Other companies are continuing to develop enhanced LLMs with the goal of creating artificial general intelligence (AGI), machines that surpass human cognition. AI firm Scale recently raised $1 billion for this purpose.
  • Akli Adjaoute, an AI expert, views AI as an effective tool but stresses its limitations in replicating human intelligence, emphasizing that AI does not understand in the way humans do, but rather follows patterns.
View Full Article

Comments (0)

Be the first to comment!