> Stepping away from this job has been one of the hardest things I have ever done, because we urgently need to figure out how to steer and control AI systems much smarter than us.
Large language models are not "smart". They do not have thought. They don't have intelligence despite the "AI" moniker, etc.
They vomit words based off very fancy statistics.
There is no path from that to "thought" and "intelligence."
Not that I disagree, but what's intelligence? How does our intelligence work? If we don't know that, how can we be so sure what does and what doesn't lead to intelligence? A little more humility is on order before whipping out the tired "LLMs are just stochastic parrots" argument.
Large language models are not "smart". They do not have thought. They don't have intelligence despite the "AI" moniker, etc.
They vomit words based off very fancy statistics.
There is no path from that to "thought" and "intelligence."