Absolutely. Mathematically more harm that good! I don't understand how people don't see it as obvious. In case you're interested, my (wordy) attempt to explain:
projectpodcast.substack.com/p/a-critical...
projectpodcast.substack.com
A Critical Re-Framing for Artificial Intelligence
It’s not Doomerism, It’s Nature: The Inevitable Reality Of an AI Future