As @zitron.bsky.social has been going of on for a bit now, I've been hurt enough times that I think I could be okay if it was a meaningful Number Go Up. But no, just financial theater and stock juicing.
There’s a distinct chance that competence will notice that they’re doing themselves no favors and not drive the car off the cliff. The incompetent won’t notice the cliff and may, in fact, continue putting their foot on the accelerator as they plummet to their death, taking as along for the ride.
It’s wild to see these tech companies still wildly throwing around investment $$ when there is a lot less VC cash floating around out there between interest rates and investments going towards infrastructure/manufacturing. Worrying for a stock market propped up by same tech companies.
idk if there are lawyers floating around in here, but do you think google actually has any liability exposure if their AI tells someone to eat a poisonous mushroom or jump off a bridge?
or are there just tons of disclaimers?
I have this exact question! Got into an argument with my husband last night because I said they could be held liable, no 230 protections for their own published words, he seems to think if they include the web sources underneath then they aren’t.
I imagine at some point the courts or Congress will have to decide if AI companies (and /or companies who integrate AI in their customer services) are liable for the info they publish. Between this and the ongoing battles over copyright /stolen content, this tech is not ready AT ALL for release
The search results underneath are not necessarily the source of the AI information though. An LLM can't really produce a source for the information it's supposedly summarizing.
And, because it's not simply linking to an obvious joke post, but instead pulling a sentence out of context and mixing it with real advice, it's essentially republishing it as their own (albeit plagiarized). Which seems like a big problem for Google if someone follows it in a bad way.
I'm curious to what extent it was actually doing summaries on the fly vs having a ton of pre-canned ones for common queries. I would have figured the latency would be too high but I suppose it's well cacheable. But in any case they don't cite the source...
Thank you! That’s what I thought, but last night when I was trying to get an AI response from Google so I could show this to hubby, I couldn’t get an AI response with any query. Right before they announced they were pulling “some” query responses. I’m still irritated he wouldn’t take my word for it😏
It most likely will come down to whether EULAs can override laws. Because most of the examples I've seen would be on the company for any deaths/illnesses caused by faulty product. Such as telling people to kill themselves or jump off bridges or eat glue.
1/
so I have no idea about any of this, and I'm sure google has a legal dept that knows more than I do, but with reddit they just say they aren't a publisher and that they're a virtual bulletin board, so if somebody posts their bomb making vid you wouldn't sue a bulletin board
2/
and with reg google it's just giving you links to click and read
but when my 10 yr old kid gets told by it's AI product that tide pods are delicious + fruit flavored I feel like that's not the same thing, regardless of how it's generated
also, beyond actual culpability, does appearance matter?
I keep testing to see if it is back up, but I haven’t gotten any AI answers from Google since about an hour before this story broke yesterday. I didn’t change any settings. Is anyone else still able to trigger an AI response?
That’s what Ive been seeing since last night too. It’s still the crappier 2024 version of Google, but I can’t get any AI responses, I’ve tried a variety of question style and nothing triggers AI- I don’t buy that they only pulled some responses
