AI development has "lost sight of the neuroscience & the basic research. It's all engineering." Karl Friston shares Demis Hassabis's concern: this "tech" engineering focus overshadows vital "science and technology" fundamentals. A call to re-balance AI research towards deeper scientific understanding, not just rapid tech advancement.

Source:

Creating content on biotech innovation and regulations can be challenging. With Tely AI, get in-depth articles automatically, packed with up-to-date research and insights to attract investors and partners. Learn more about Tely AI.

Its a really good point. There are fundamental things about the essence of life being overlooked in the engineergasm that is happening around AI.

Surprised by the lack of AI use in moral philosophy. In an AI-everywhere future, what human life is about beyond the intelligence part feels more important than beating coding benchmarks or testing 13-digit multiplication and then embed more productively for human betterment...

It's would be a good advice if we had decades woth no real progress. "You're stuck because your methodology is too far from science". Yeah, maybe in some parallel universe with AI winter that would make sense. But not in a universe where AI is the fastest developing science.

Thanks for posting this.

Too slow, they are just afraid to implement the Loom that’s in the room already

There is a solution from cognitive science that has been ignored by computer sciences’ obsession with scaling and engineering. @jbthinking shows the path forward when implementation, ie. engineering is led by the right science. Spoiler alert: the brain is not a processing engine like a computer

As soon as the $$$ approach runs out of steam, then there will be time for more fundamental research. With which AI will certainly help

AI's future lies in harmonizing engineering with neuroscience. Let's not lose the 'science' in 'science and technology.'

Absolutely agree, science and technology should go hand in hand for AI progress.