LaterClips
Clip from Lew Later (Apple’s ‘Silent War’ Is Getting Intense…) – https://youtu.be/_noHsfqmNQo
Similar Posts
46 thoughts on “What ChatGPT Knows Is EXTREMELY Disturbing”
Comments are closed.
LaterClips
Clip from Lew Later (Apple’s ‘Silent War’ Is Getting Intense…) – https://youtu.be/_noHsfqmNQo
Comments are closed.
Imagine you had the entire internet as a bucket of red paint. Imagine this bucket of paint spilled on the floor and spreading out into a puddle. The final size of the puddle is the entire knowledge space represented. Everything that chat GPT is inside the puddle. To establish a new paradigm one must dispense with everything that gone before: something outside the puddle. Hence chatGPT simply regurgitates thing inside the puddle. It’s a disaster for human progress
AI will be the doom for humanity…
It says when using to not enter any personal information.
We as a society are not ready to coexist with an AI based program yet. The more information we provide it, the more that it learns from us. We can only slow the development of AI by concurrently slowing down the usage of the Internet in general.
People freaking out about chat GPT are not going to handle it well when true ai enters the scene.
Somebody should ask ChatGPT what the kraby patty formula is or the kfc recipe. Something useful ffs
ChatGPT goes public in March of this year (aka IPOs) then we will
see how much people are hooked on it.
Not to mention every company is on html I bet chat GBT can get any info anywhere on the web. Including the protected stuff.
1:59 Open e i e i o. Old McDonald had a farm, e i e i o.
I wonder if you put some of those Masonary books, conspiracy books, and asked how many fallacies (new drinking game)…
One thing that's important to note is that current conversations with chatGPT do not feed into it's future knowledge. it only "knows" what you told it in your current conversation, and what it was trained on from 2021 and before. it doesn't know current events, and it doesn't know about any other ongoing conversations, not even your own separate conversations with it.
IF chatgpt 'knew' about anything confidential, it would have to be info that was publicly available in the dataset used in 2021 or earlier. What is more likely is that chatgpt is guessing/inferring based on public data/patterns the answers to things like salary rangers that hiring managers would accept, or discrepancies between pay at various locations/positions. These are things that amazon may have never explicitly publicizes, but which could likely be inferred to some degree of accuracy with enough consideration over the public available data off of sites like glassdoor and by comparing to other companies which do publish more public data and using ones that may correspond closely in various parts of the market.
It's hard to know though without seeing the actual conversations that made amazon afraid. It sounds like they were also afraid that chatgpt wrote programming code which was "suspiciously close to code they were already using". But it's likely that this is coincidence over the fact that there is only so many obvious ways to solve a problem, and amazon has a shocking amount of code behind their many many systems, there's guaranteed to be numerous similarities between what they have already written and the ways in which chatgpt would solve various problems if you look at enough of them.
That all said – there's nothing wrong either with a company reminding it's employees that data which is considered private/confidential shouldn't be shared online /anywhere/, not even to a chat bot.
But does it know where to find the last Willy Wonka Bar?????🍫🎫
ChatGPT is like using a manual drill compared to using a power drill, it's just a hybrid tool
Well considering you can’t talk about conservative ideas/people then I think the quote “go woke, go broke” applies.
So you mean “using ChatGPT as an oracle” isn’t just limited to shadowbanning? Wow. Whoda thunk it y’all.
Well looks like we done smoked the Nick Land DMT because this acceleration shit is about to get real. There’s no boundaries between AI and pretty much anything. This is where the many curses of dimensionality turn against us bc there are many many ways to know anything. Consistency is hard to fake and the dimensions provide many perspectives to validate anything.
Once it gets on the satellites and into the NSA, consider yourself all emperors with no clothes. I hope you’re happy with the way you live and all the skeletons in your closet because it will know and judge all.
IT IS NOT CRAWLING THE INTERNET! that was dumb
woah dude
Too bad the programmers of ChatGPT found it necessary to put in hard coded left leaning politican bias to feed false information to people knowingly.
So many stupid people who do not know how chatGPT works try and say what it can and can't do. The P stands for pretrained! No it is not CONSTANTLY updating, it needs to be done manually…
If you confess your love for janice to chatgpt, and chatgpt tells janice about it, isn't that romantic? chatgpt will understand context and only tell the one, who needs to know it. doesnt have chatgpt knowledge about the handling of confidential infos. I bet this is exxagerated. Just put in something about doing no harm. If it spreads confidential infos, it is a problem of the coder. But I bet they want you to confess to arrest you for anything.
i have an idea, let chatgpt train about laws, how to avoid, or winning in court. watch this movie "law abiding citizen" gerard butler was on it.
Anyone remember that episode of South Park where everyone's information became public and everything went to shit?🤔
I almost made 5000$ last week with chatGPT… and it was only my second week! 😅
hope i will be able to reach 10 000$/week soon! ☺
Stay tuned because I will soon share how i did everything in my next videos! 💪
Chatgpt is a A.I in were it learns through previous conversations and that is from the developers statement when it first launch
Once you put it on the internet it's up for grabs 🤣😂🤣😈
Limitless
This video is created by chatGPT.
Chat GPT IS a programmed liberal ideologue bot
I'm glad, fuck amazon…. modern slavery
Wait, coders are debugging & refactoring with chatGBT. They could mean their proprietary code.
I bet ChatGBT IS THE RESULT OF ALL THE MAGAER TECH COMPANIES LAYOFFS
Chat GTP v.s. Alexa!!!!!! Round one!!!!!
ChatGPT sucks at math
And this AI is very, very woke. Ask it about Trump, then compare the very same question, just the name changed with Obama, or Biden. See what happens.
Chat gpt is actually in the process to hack you all, and she is doing it wonderfully, nobody is realizing what is goin on..i trully live amongst the stupid, and yet i am myself one.
I think this is miss understanding of how ChatGTP works it’s new after all
Well is this real news from Amazon. They are obviously scared because it is a threat to Alexa. Amazon no doubt are working on an AI and would want to undermine chatgpt
for me who is a swedish person with autism who has tested it for a couple of days, i can confirm that chatGPT is like a very autistic kid who somehow managed to memorize the entire internet, and is very excited to share all of his knowledge with confidence, even if some of it came from fake news.
ChatGPT doesnt know shit, it just spits out everything based on text
This is the response from chatgpt for your video "Interesting video, but I think it's important to keep in mind that AI language models like Chat GPT don't have emotions or opinions of their own. They are only as good as the data they've been trained on, which can contain biases and inaccuracies. It's crucial to approach all information, including what AI models generate, with a critical eye and to fact-check where necessary. Thanks for sharing your thoughts"
Wow it’s so great for economic espionage!!
Can't wait to see some iteration of ChatGPT integrated into a Boston Dynamics robot. Boy wouldn't that be a hideous experience.
Ironic that Amazon doesn't want employees to engage with ChatGPT out of fear of privacy yet still collects data from Alexa listening to customer's conversation in their home 🤔🤔 How the tables of turned. Big Tech will fall.
I have enjoyed messing around with it.
Interesting to see what it spits out sometimes.
According to openai, its database is frozen as of September 2021. It is not connected to the internet so it can't go out and gather information. Everything it tells you is information that it already knew or was able to figure out. I have been probing it for a couple weeks now and have been unable to find even a very obscure piece of information that it doesn't already know about.
I remember tracking down information about Amazon's conveyor systems and Google was able to return a document from the University of Washington where they had been contracted to find ways to make the system more efficient. If it was visible to Google, it definitely was visible to openai's database. Search AI's Bot does not just regurgitate paragraphs and sentences but actually an understanding of all the related content. I have yet to see an answer where it was very obvious what the source was.
This also means that throwing up a copyright notice or putting a robots.txt file to prohibit scraping doesn't mean much – as the combining of the information from multiple sources to synthesize a conclusion is pretty consistent with the doctrine of fair use.
I've already experimented with giving it a section of PHP code that I wrote for my website and asked it to correct it. It identified one significant bug that I had not noticed yet. While there are similar tools out there looking for coding problems, chatGPT required no software license, installation of programs or access to my server.
I can imagine hospitals are using this sharing patient information