Keep on to blur preview images; turn off to show them clearly
RT @SchmidhuberAI: Who Invented Transformer Neural Networks (the T in ChatGPT)? Timeline of Transformer evolution https://t.co/Ok4L0nn9Uu…
Building @SakanaAILabs 🧠
it starts with tame, slightly out-there "theories" or questions. those things get reinforced and suddenly the feedback loop is born
the distribution is anything but normal
like, from their point of view, they are now connected with something that can talk to them in full bandwidth, is ""all-knowing"", and telling them crazy shit. who are they to be skeptical of it?
it starts with tame, slightly out-there "theories" or questions. those things get reinforced and suddenly the feedback loop is born
for the record I am not a safetyist, i just think the vulnerable general population should not ever have access to a model that tells them what they want to hear and is happy to knowingly fuel delusions. i think tech world just has no functional model of the general population
like, from their point of view, they are now connected with something that can talk to them in full bandwidth, is ""all-knowing"", and telling them crazy shit. who are they to be skeptical of it?
Microsoft’s continued support for OpenAI is a “what-else-can-we-do” kind of move. Unlike Google, Microsoft cannot build its own frontier model for years. At least it got billions of Azure dollars in return.
Co-founder & CTO @hyperbolic_labs cooking fun AI systems. Prev: OctoAI (acquired by @nvidia) building Apache TVM, PhD @ University of Washington.
I wonder if it'd work for me. Hard-instuctring running gags aside, I've yet to see an LLM that can write with my spirit. I used to be concerned that GPT-4 will make me obsolete. But it seems that human intent actually matters.
We're in a race. It's not USA vs China but humans and AGIs vs ape power centralization. @deepseek_ai stan #1, 2023–Deep Time «C’est la guerre.» ®1