BBC
Grok is part-smart, part-dumb (so is Elon Musk)
If you have time to read the article, this is an interesting insight into the limitations of current AI models, and what the next step - or a different way of doing things - could be. It puts current debates on AI in perspective.
AI-powered humanoid robots are not very good at doing fiddly things that a human being, once properly trained, can do easily. This would include simple tasks such as filling up the dishwasher. Trades such as plumbing are not going to be replaced by machines any time soon.
The Large Language Models (LLM), on the other hand, are very good at collecting information, analyzing it, and summarizing it: this is what many white-collar jobs entail in sectors such as banking, insurance, logistics, administration, etc. So, at this point, this new industrial revolution is going to hit white collars frontally, when at least some trained blue collars will be spared, paradoxically.
At the end of the day, even sophisticated AI machines are just intelligent parrots. They regurgitate data they have found online. There is nothing actually creative in what they come up with: if they mention it, it's because they've found it online, and that's because it was there already, generated by human experience. They are not genuinely capable of critical analysis, let alone emotional empathy. But they are very good at packaging a summary with bullet points that gives you a quick overview of an issue.
In certain fields, however, given the enormous amount of data the machine is able to collect and analyze, they have already been able to come up with genuinely new ideas. E.g.: finding new drugs that no one had thought of developing, based on the analysis of known substances and suggesting that they be applied to an existing medical problem that did not have a solution.
_______________
"We don't have robots that are nearly as good at understanding the physical world as a rat," says Yann LeCun, one of the leading figures in the world of artificial intelligence.
He worked at Facebook-owner, Meta, for a decade, where he was chief AI scientist, but left in 2025 and founded Advanced Machine Intelligence Labs (AMI Labs).
His goal is to move AI beyond current systems like ChatGPT, Claude and Gemini. They have their uses, he says, but will never be able to tackle complicated situations in the real world, like getting a robot to do household chores. [...]
Large Language Models (LLMs) like ChatGPT are extremely good at some things like coding, mathematical problems and generating text, LeCun says.
But he argues that these are well defined and predictable problems.
"They [LLMs] basically just accumulate knowledge... They can regurgitate something, you train them to regurgitate, but they're not particularly smart. They don't have an underlying understanding," he says.
In the real world there is a bewildering array of outcomes to any action, which requires a more flexible type of artificial intelligence. [...]
What will happen to humans in a world where robots can operate independently?
"We're still going to need humans to figure out what questions to ask, what to build, what to create, which is really the properly human aspect," he says.
The AI will work for us he adds.
"Our interaction with future AI systems - even if they are smarter than us - is going to be like the interaction between a captain of industry or a political leader with their staff of assistants - many of whom are smarter than they are."
