The other major vocabulary/semantic shift is that we're comfortable calling applied ML -> AI. Until recently, you would always say ML for applied products, AI w

The other major vocabulary/semantic shift is that we're comfortable calling applied ML -> AI. Until recently, you would always say ML for applied products, AI was either research or fraud/exaggeration or very least meant you had some futuristic vision, you wouldn't use it for things you were generally applying right now in your product. I technically had a "focus in AI" in my grad degree and even my advisor said to label it as "ML" in my resume otherwise experts would think I was just a fraudster or not knowledgeable at least (~10 years ago).