Raw LLM Responses

The ethical and sustainable part is also something that's been criticized but never really brought up in discussions—I think the amount of money being pumped into AI related projects allows people to conveniently not care about that. >I know we won't hit the promises that AI vendors make This is one of the main take aways from that MIT report—most AI projects don't take off, but to be fair, most IT projects in general also don't take off. The difference here is the huge amount of hype and expectation tied to the AI projects. Edit: just to expand on that since people keep saying I didn't read the report—the report itself doesn't explicitly state that "AI vendors" overpromise or anything like that, here is an excerpt from the link (bold mine): >generative AI implementation is falling short. “The **95% failure rate for enterprise AI solutions** represents the clearest manifestation of the GenAI Divide,” the report states. The core issue? **Not the quality of the AI models**, but the “learning gap” for both tools and organizations. While executives often blame regulation or model performance, MIT’s research points to **flawed enterprise integration.** I want to reiterate that I never said AI models are shit—in fact I use them daily, it's the constant hype and wanting to shove it into everything that bothers me. It reminds me of blockchain and crypto—eventually the noise will settle, I hope.