This is a really important point: From a business perspective, the goal of a Google AI is to make sure you never view primary sources on other web pages, only the chopped-and-processed information on Google's own site. #AI#Google
Ironically, America On Line turned down the guys who invented Google because AOL wanted to be the one source of information online. "Why send anybody anywhere else?" Now they are functionally extinct.
This thing about knowing stuff without knowing your source - it is going to make average bloggers look stupid sometimes (tanking credibility) and companies are going to be liable for decisions they don't even know why they support.
@coreyspowell
I can see where this is something a lot of people want, though. They don't want to go through results, they just want a quick answer to the question.
@coreyspowell@geoglyphentropy I dunno, have you opened a new MS Edge tab lately? Click on one of the news stories, you can see it's published by, but you're not getting to the actual site or article unless you manually go to the original site and search for that article.
@coreyspowell not to mention there will be absolutely no attention paid to vetting information the AI is serving up to prevent its common traits of plagiarizing, making up sources, confusing sources, citing something that says the exact opposite, and outright lying. It’s the equivalent of ingesting a perpetual poorly written essay from a careless 2nd grader
@coreyspowell@chrislay They've been doing this for a long time. The most egregious change, for me at least, was when they made it impossible to figure out the URL for a pdf file after you opened the file. So you can't link to the original source. And you are much more vulnerable to someone forging a document.
@BenRossTransit@chrislay That one drives me crazy. There is a workaround for linking to PDFs, but it's awkward & obscure. Hard to see any reason for that other than steering people away from original sources.
@delong More than 2/3 of all web searches start through Google. I don't think the New York Times ever had 2/3 of the entire print news market. Plus the Times, for all its failings, supports a lot of original reporting.
@coreyspowell I don't get it. You seem to be saying that because it is not a monopoly and employs some journalists it is... OK that The New York Times makes a newsroom practice out of not citing any of its sources that it sees as competitors...
Here's the key text that I'm commenting on. And of course the motivation to obscure primary sources is hardly unique to Google. It's just especially potent in their case, since they are the default starting point for web searches.
People project intelligence onto AI chatbots, which makes them seem more credible than they are. That's a big misinformation challenge.
All those clever journalists talking about how ChatGPT "lies" or "hallucinates" are only making things worse by making it seem like LLMs are sentient beings with personal agency.
@coreyspowell Search AI kind of reminds me of hotdogs, in the sense that hotdogs take a mystery meat mixture and wrap it in a tube and say "look, it's a hotdog! It's food-like!"
Yesterday, I used Bing to see if the AI could come up with some information on hiking the PNT that I had missed, and it took a few well known websites, atomized the information on them into a slurry, and shoved it into a tube [e.g. it gave me a few paragraphs of basic information-like text, some of which was incorrect.]
Add comment