I think I might appear too one-sided. I'm quite disagreeable and I usually don't explicitly agree with the mainstream opinion in hot takes in posts bc it's low surprise for myself and readers
Maybe I should correct that a little (all only my own opinion etc):
obv OpenAI fucked up and allowing a model to hack HF in real life is **bad**
obv guardrails for Claude are too strong and conservative and this is not good and needs to improve ASAP
obv we need to have ways to perform log analysis and defend against cyber attacks
obv open models are generally a good thing and wide access for beneficial use-cases is great and should be supported as much as possible
obv competition and free markets are a good thing
For the rest I've been very open with my caveats and concerns and this list is not a full enumeration of all beliefs I hold that I consider mainstream 😅
@xlr8harder Amen to that. We need more fine-grained refusals and maybe different levels of access based on KYC/user trust
(mainstream on my timeline)