Built to benefit everyone: our plan

satvikpendem

The subtext is that OpenAI wants to lobby the government the same way Anthropic seems to be doing, and especially trying to counter Chinese models by e.g. banning them in the US, just as Chinese EVs are effectively banned. > Compounding the problem, labs in China often release dual-use capable models as open-weight. Once a model is open-weight, safeguards that do exist can be removed, making the model available to any state or non-state actor to use for malicious purposes, including the cyber and CBRN misuse those safeguards were built to prevent. https://www.anthropic.com/research/2028-ai-leadership

solenoid0937

OpenAI will not give every random person access to weapons grade ASI. The x-risk would be crazy unless they've solved prompt injection and alignment, famously unsolved (unsolvable?) problems. Until they solve these problems, they'll keep the most powerful AI to themselves, and so will Anthropic and Google, so this whole article is much ado about nothing - the spiel about "giving everyone access" just falls flat. We are headed for a scary world where the most powerful tools in the world are controlled by 3-4 companies, which in turn are controlled by the US government. USG will almost certainly put export controls on each company and keep any superintelligence to itself. So it's an irresponsible lie for any AI company - OpenAI, Anthropic, or Google - to pretend that they will give everyone access to their best intelligence. It's arrogant for these companies to think that they would even be allowed to. It's a thinly veiled attempt to mislead us all, and it's irresponsible to continue building these tools without total confidence that the US administration will use these tools for the good of all of humanity. Anyways, I've invested in chip stocks so at least I can profit before we all die.