I think the problem is that tech companies have overstated the benefits of AI and for obvious reasons they have been understating the risks associated with rel

I think the problem is that tech companies have overstated the benefits of AI and for obvious reasons they have been understating the risks associated with relying on it. There are examples in the article dating back to 2015 but for non-tech people today, it might come across as "If an AI can converse, draw pictures of what they ask and identify a bunch of random things, it can surely identify a mere mushroom"