News for hackers. Stuff that matters.

A comment on Gemini: Google's most capable AI model yet. Parent comment

Re:Gemini: Google's most capable AI model yet by ithkuil on Wednesday December 06, 2023 @06:20PM (#38550834)
Shouldn't be hard to just say so instead of claiming it's multilingual readiness?
[ Parent | on HN ]
Re:Gemini: Google's most capable AI model yet by throwaway09223 on Wednesday December 06, 2023 @06:55PM (#38551116)
Quite the opposite - it's not only hard it's also unwise.

Admitting that you know that your product may create legal liabilities is not a very smart thing to do.

[ Parent | on HN ]
Re:Gemini: Google's most capable AI model yet by paulryanrogers on Wednesday December 06, 2023 @07:04PM (#38551180)
Did OpenAI ask for permission or forgiveness in that regard?
[ Parent | on HN ]
Re:Gemini: Google's most capable AI model yet by leptons on Wednesday December 06, 2023 @07:28PM (#38551349)
Did OpenAI board ask for permission or forgiveness? Seems like it was forgiveness and they lost.
[ Parent | on HN ]
Re:Gemini: Google's most capable AI model yet by paulryanrogers on Thursday December 07, 2023 @01:43PM (#38560079)
If boards cannot hold CEOs accountable then who can?
[ Parent | on HN ]
Re:Gemini: Google's most capable AI model yet by leptons on Thursday December 07, 2023 @05:10PM (#38562517)
Except their CEO wasn't held accountable, it was a coup, and the board was removed and the CEO reinstated.
[ Parent | on HN ]
Re:Gemini: Google's most capable AI model yet by paulryanrogers on Thursday December 07, 2023 @05:51PM (#38562898)
Right, so then, who is left to hold CEOs accountable?
[ Parent | on HN ]
Re:Gemini: Google's most capable AI model yet by rvnx on Thursday December 07, 2023 @12:07AM (#38553051)
They could create a separate limited liability legal entity that is a full subsidiary of Google. That subsidiary could license the technology or platform from the parent company and they'd be able to launch worldwide.
[ Parent | on HN ]
Re:Gemini: Google's most capable AI model yet by throwaway09223 on Thursday December 07, 2023 @12:49AM (#38553289)
This could be an answer to an intent to violate any law.

The counterpoint is that this doesn't actually work in practice.

[ Parent | on HN ]
Re:Gemini: Google's most capable AI model yet by rvnx on Thursday December 07, 2023 @02:04AM (#38553676)
Well that's Deepmind I guess ?
[ Parent | on HN ]
Re:Gemini: Google's most capable AI model yet by ithkuil on Thursday December 07, 2023 @02:32AM (#38553834)
You can spin it in a different way:

"We committed to meet all the regulation in each country we're operating in. Our teams are working on it. In the meantime we'll start in our domestic market, because obviously if there has to be a first country to open it would be our origin country, right?"

[ Parent | on HN ]