AI is rife with scams and hype, says top Google official

According to Demis Hassabis, AI is attracting too much investment, which could put a brake on its primary goal of scientific advancement

Demis Hassabis, CEO of Google DeepMind, has said that large waves of investment in AI could cloud the outlook by attracting potential scams and massive hype. He believes that the $42.5 billion raised in more than 2,500 funding rounds during 2023 could play a trick on the future of AI.

DeepMind was founded in 2010 by Hassabis. Its main goal was to create an artificial intelligence that matched human cognitive ability and use it as the ultimate tool of science.

The obsession with scientific advancement was to blame for DeepMind’s success, a stance that Demis says is being clouded by money and overfunding.

AI’s extreme funding may hold back scientific progress

The huge avalanche of money that has buried the AI environment over the past few years gives Demis a controversial opinion, «they bring a lot of hype and maybe some scamming». He also alludes to the crypto world in a negative tone, which, he says, can bring darkness into the picture.

The launch of the ChatGTP chatbot in 2022 sparked a wave of investment for start-ups to focus on creating generative AI. The global interest in the technology has shifted away from promoting scientific breakthroughs to other areas.

Hassabis believes that the promotion of AI is moving towards a landscape of unrealistic propositions, leaving behind its true scientific goal, which he believes could be the most important breakthrough in history.

The golden age of scientific discovery?

«We are at the beginning, perhaps, of a new golden age of scientific discovery,» Demis told the Financial Times. According to him, AI is currently only scratching the surface of scientific discovery over the next decade. His goal has always been for artificial intelligence to be the ultimate tool of science.

His track record includes good examples, such as AlphaFold, a DeepMind model launched in 2021 that was able to predict the structures of 200 million proteins and is now used by more than a million biologists worldwide.

Hassabis sees AGI (Artificial General Intelligence) for the next decade

Talking about the future of AI, Demis spoke about AGI or Artificial General Intelligence. This type of AI is hypothetical and bases its development on matching or surpassing human intelligence. He believes it is paramount to provide a greater scientific focus to develop AGI faster because of its importance in breakthroughs.

«I wouldn’t be surprised if it (AGI) happens in the next decade,» the DeepMind CEO added to his statements.

Hi I’m Anastasia and I write about AI tips, tricks and tools, especially Midjourney !
If you find this article useful and would like to support me :
-👏🏻 Give me a Clap to this article.
-🎨Follow me for more interesting content.