If these conditions were on an "open source" model I'd have a fit because I fundamentally disagree with the way companies are trying to enforce how model weights they release are used while capitalizing on open source branding.
They do however all seem pretty reasonable and uncontroversial (unless I missed something) and I understand why AWS or others would want to enforce these conditions. The dividing line for me would be if use restrictions drift into the realm of politics (which "disinformation" potentially could).
I'd rather see well thought out regulation govern uses and not have monopoly providers writing the rules.
agree - well said. Also here is another simple aspect.. a ChatGPT transcript recently showed "as an AI, I do not have access to library collections or external databases" ... several times in a session on certain technical topics. So obviously, hooking up to actual library services for document retrieval, or API calls for actions.. is a giant competitive topic, and not-less a safety topic.
"to perform a lethal function in a weapon without human authorization or control."
In other words... Just make sure you have a confirm dialog before launching the rockets!
IMO the major cloud players have no ethical or moral controls in place that are actually practical or useful in curtailing the usage of AI methods for the purposes of war or surveillance. It's inevitable at this point. The OpenAI/Microsoft partnership will inevitably lead to contracts with NSA/CIA/DoD. AWS/Microsoft have no real ethical pushback on DoD relationships. Google at least seems to have some internal political opposition and moral compass. The future will be a weird place.
7 comments
[ 3.0 ms ] story [ 38.6 ms ] threadThey do however all seem pretty reasonable and uncontroversial (unless I missed something) and I understand why AWS or others would want to enforce these conditions. The dividing line for me would be if use restrictions drift into the realm of politics (which "disinformation" potentially could).
I'd rather see well thought out regulation govern uses and not have monopoly providers writing the rules.
In other words... Just make sure you have a confirm dialog before launching the rockets!
IMO the major cloud players have no ethical or moral controls in place that are actually practical or useful in curtailing the usage of AI methods for the purposes of war or surveillance. It's inevitable at this point. The OpenAI/Microsoft partnership will inevitably lead to contracts with NSA/CIA/DoD. AWS/Microsoft have no real ethical pushback on DoD relationships. Google at least seems to have some internal political opposition and moral compass. The future will be a weird place.
"I Quit Google Over Its Censored Chinese Search Engine. The Company Needs to Clarify Its Position on Human Rights" - https://theintercept.com/2018/12/01/google-china-censorship-...
You make a gun and say it’s for responsible use only but sell it anyway without any checks and balances.
Which congress folks are gonna get lobbied the most for providing more immunity?