Meta CEO Nick Clegg says current artificial intelligence (AI) models are "pretty stupid" by minimizing the potential risks of using them.
He said the excitement of the emergence of artificial intelligence was premature compared to the degree of its technological development. Clegg explained that current AI models are far from thinking on their own, but are "in many ways quite stupid," said a representative of the company that owns Facebook and Instagram.
Meta has announced the general free use of the language model "AI Llama 2" even for commercial purposes.
Clegg said that such platforms that run "chat bots" like ChatGPT, basically just merge parts of a huge language base and make sentences by guessing what the next word in the row would be.
Clegg added that the warnings of an "existential threat" to humanity by some AI experts refer to systems that do not yet exist.