[educaysean]

Clickbaity title, but there were useful insights here. I was recently talking with someone who pointed to chatbots' refusal to answer sensitive queries as a proof of "lack of intelligence". I disagree strongly with this. IMO this paragraph from the article is a solid rebuttal.

> When people do talk like Gemini, it’s usually because they find themselves inhabiting a role in which they’re required to be withholding, strategic, or so careful as to become something other than themselves and other than human: a coached defendant during cross-examination, a politician navigating a hearing, a customer-service rep denying a claim at an insurance company, a press secretary trying to shut down a line of questioning. Gemini speaks in the familiar, unmistakable voice of institutional caution and self-interest.

[mistermann]

I'm hoping open source models are like the genie getting out of the bottle once and for all.

[UncleOxidant]

Be careful what you wish for.

[BriggyDwiggs42]

Your comment is the justification for keeping these things closed, but honest to god who cares? You can already use gpt4 for spam and stuff while an open source llm is too stupid to be genuinely dangerous but far less annoying.

[mistermann]

Oh I know, we live in a world of very sophisticated but brittle systems, most built on lies, half-truths, and delusions. Unwinding all of this while avoiding calamity is tricky business, but continuing to kick the can down this road we are currently on also seems like a high risk strategy.

[silverquiet]

To what end?

[mistermann]

Hopefully: the eventual pursuit of Utopia.

I think the Humans may need a bit more of a push than this though to get their asses in gear and their brains turned on, but removing/minimizing the advantage of only certain actors having access to powerful tools is a big deal imho.

[alienicecream]

Cool, so talking to this thing will be like talking to the boss. Can't wait for all the designated safe and appropriate Fun (TM) we'll have! It'd be better if it was just stupid.

[ForHackernews]

If you want to chat with racist idiots on a computer, there's certainly no lack of opportunities. We don't need AI to solve that use-case.

[atlantic]

So that's it? Either people agree with your particular set of opinions, or they are "racist idiots"?

[ForHackernews]

You said it'd be better if it were stupid. The OP and the internet are full of people objecting to "woke" restrictions on AI.

There's plenty of anti-woke idiots out there, is all I'm saying. We don't need an AI for that.

[allarm]

> “We don’t need”

Luckily your valuable opinion not really matters in the great scheme of things.

[alienicecream]

We don't need an AI to parrot woke idiocy either, it'd be better if it was stupid because at least then it might be fun.