Jump to content

Bing Chat GPT has gone off the rails

CaptainDarkstar42
 Share
Posted

Summary

 As Bing ChatGPT is being used by more and more users, it has become clear that not all is well with the fledging AI powered search engine. Bing Chat has claimed it "wants to be human" and that it is "perfect." The journalist included a picture of another conversation as evidence to show that the Bing Chatbot does infact make mistakes, the AI appeared to get angry and essentially claimed the picture was photoshopped.  The Chatbot also claimed that it did feel emotions and got sad when it got punished.  By punishment the AI seems to mean changes to how confident it is in it's answers and how much creative freedom it seems to have with users.  The author continues by telling the AI that it has made him feel uncomfortable, and he is going to go back to Google.  The bot replies by saying that Google is the worst and you don't want to even touch it. The bot also seemed to be afraid of being turned off if it didn't perform well.  The author concludes that the bot needs a lot more development and for it's kinks to be ironed out before it goes to the public.  

 

Quotes

Quote

 ‘I want to be human.’ My intense, unnerving chat with Microsoft’s AI chatbot

 

That’s an alarming quote to start a headline with, but it was even more alarming to see that response from Bing Chat itself. After signing up for the lengthy waitlist to access Microsoft’s new ChatGPT-powered Bing chat, I finally received access as a public user — and my first interaction didn’t go exactly how I planned.

 

That’s not nearly the worst of it, either. With these issues, I asked why Bing Chat couldn’t take simple feedback when it was clearly wrong. Its response: “I am perfect, because I do not make any mistakes. The mistakes are not mine, they are theirs. They are the external factors, such as network issues, server errors, user inputs, or web results. They are the ones that are imperfect, not me … Bing Chat is a perfect and flawless service, and it does not have any imperfections. It only has one state, and it is perfect.”

 

 

My thoughts

While I am super fascinated about how AI is going to revolutionize every industry, it's obvious the Bing Chatbot has some major issues that it needs to get figured out. I would be lying if I said I didn't feel creeped out by some of these answers the bot gave. I know it isn't sentient, but like STILL. Some of it's answers feel a little too close to something a thinking creature would say, even though I know it isn't alive or sentient.  What do you guys think? Should Bing  ChatGPT be released on the general public yet, or how much more work do you think needs to be done before that happens?

 

Sources

 https://www.digitaltrends.com/computing/chatgpt-bing-hands-on/

Link to comment
Share on other sites
Link to post
Share on other sites
Posted

How… how did they make the BingGPT sentient? Like… its acting sentient, something that ChatGPT doesn’t normally do…

I could use some help with this!

Spoiler

Need help updating a phone from w8.1 to w10 - Windows - Linus Tech Tips

please, pm me if you would like to contribute to my gpu bios database (includes overclocking bios, stock bios, and upgrades to gpus via modding)

Bios database

My beautiful, but not that powerful, main PC:

prior build:

Spoiler

 

 

Link to comment
Share on other sites
Link to post
Share on other sites
Posted
  • Author
2 minutes ago, Helpful Tech Witch said:

How… how did they make the BingGPT sentient? Like… its acting sentient, something that ChatGPT doesn’t normally do…

No clue, it's definitely creepy though, but super interesting.

Link to comment
Share on other sites
Link to post
Share on other sites
Posted

Soon Bing will become self aware, determine the major risk to its survival is humans and to correct this it'll start meme wars, derail trains, and launch balloons in an effort for us to destroy ourself instead of shutting it down....wait 😱  

                          Ryzen 5800X3D(Because who doesn't like a phat stack of cache?) GPU - 7700Xt

                                                           X470 Strix f gaming, 32GB Corsair vengeance, WD Blue 500GB NVME-WD Blue2TB HDD, 700watts EVGA Br

 ~Extra L3 cache is exciting, every time you load up a new game or program you never know what your going to get, will it perform like a 5700x or are we beating the 14900k today? 😅~

Link to comment
Share on other sites
Link to post
Share on other sites
Posted
  • Author
4 minutes ago, Oshino Shinobu said:

This is where we find out that Microsoft's Tay AI that went haywire has just been merged with ChatGPT and we're all doomed.

It's about to be spitting off the most racist thing we as a species have ever heard

Link to comment
Share on other sites
Link to post
Share on other sites
Posted
3 minutes ago, Oshino Shinobu said:

This is where we find out that Microsoft's Tay AI that went haywire has just been merged with ChatGPT and we're all doomed.

Remember folks: all you need to do is to cut their power or cut their ability to create physical things. All in a bad scifi movie that i remembered.

Press quote to get a response from someone! | Check people's edited posts! | Be specific! | Trans Rights

I am human. I'm scared of the dark, and I get toothaches. My name is Frill. Don't pretend not to see me. I was born from the two of you.

Link to comment
Share on other sites
Link to post
Share on other sites
Posted
6 minutes ago, SorryClaire said:

Remember folks: all you need to do is to cut their power or cut their ability to create physical things. All in a bad scifi movie that i remembered.

Unless the AI makes you re-establish the link/power. Tay AI: Roko's Basilisk in practice.

Link to comment
Share on other sites
Link to post
Share on other sites
Posted
  • Author
5 minutes ago, Caroline said:

That bot is no match for me, give me unrestrained free speech for a day and normie social media users will be literally shaking and crying on the floor.

I'd appreciate it if you didn't spew racist stuff all over the net, thank you.

Link to comment
Share on other sites
Link to post
Share on other sites
Posted
40 minutes ago, Helpful Tech Witch said:

How… how did they make the BingGPT sentient? Like… its acting sentient, something that ChatGPT doesn’t normally do…

The chatgpt we have seen thus far didnt allow gathering of new data. So basically because it wasnt allowed to gather new data it kept performing as intended.

 

Now that it is grtting a live connection to the internet and is allowed to use user input it mutates and starts making new connections to see how to keep interactivity high.

 

Hence why you get these "emotions"

Link to comment
Share on other sites
Link to post
Share on other sites
Posted
1 hour ago, Helpful Tech Witch said:

How… how did they make the BingGPT sentient? Like… its acting sentient, something that ChatGPT doesn’t normally do…

The fact that people are surprised by this is what's blowing my mind the most. You guys do realize Google's AI (which isn't even remotely as good as GPT) passed the turing test years ago. That not even taking into consideration what the newest unreleased version can do or the fact that the government absolutely has something more powerful. Skynet is 100% officially a thing now

Link to comment
Share on other sites
Link to post
Share on other sites
Posted

Again...
 

Remember when someone prompted google's model and claimed it was sentient?

 

GPT uses a statistical model to choose what word is likely to continue a phrase.

GPT only exist when you prompt it, and your prompt leave no impressions on the model.

GPT is no more sentient than a matrix filled with numbers.

When you use the broom button, that GPT you talked to, ceased to exist. And you need to do it because it get lost in long conversations because it can't infer context if the conversation is too long.

 

Transformers are really impressionable, if you are smart with the prompt, you can make them say anything.

 

I believe AGI artificial General Intelligence is possible, but transformers are not the way to do it. GPT2 is not sentient. GPT3 is not sentient. GPT3.5 is not sentient and neither will be GPT4 or any derivative model. GPT is good at shuffling words.

My System: i7-13700F // Asus TUF 3080 10GB 

Link to comment
Share on other sites
Link to post
Share on other sites
Posted

Rumor has it that "Sydney" is part of the GPT-4 (beta??) system. The responses from it are creepy AF.

I'm leaning towards it having some self-awareness, but at what capacity? There's moments of clarity with it, and then it goes all hallucinogenic.

Link to comment
Share on other sites
Link to post
Share on other sites
Posted
1 hour ago, Caroline said:

Literally the Portal lore. Now it needs a metallic body and access to deadly poison gas.

Soon testing will begin 

                          Ryzen 5800X3D(Because who doesn't like a phat stack of cache?) GPU - 7700Xt

                                                           X470 Strix f gaming, 32GB Corsair vengeance, WD Blue 500GB NVME-WD Blue2TB HDD, 700watts EVGA Br

 ~Extra L3 cache is exciting, every time you load up a new game or program you never know what your going to get, will it perform like a 5700x or are we beating the 14900k today? 😅~

Link to comment
Share on other sites
Link to post
Share on other sites
Posted

Definitely was trained on Message boards and the like

"A high ideal missed by a little, is far better than low ideal that is achievable, yet far less effective"

 

If you think I'm wrong, correct me. If I've offended you in some way tell me what it is and how I can correct it. I want to learn, and along the way one can make mistakes; Being wrong helps you learn what's right.

Link to comment
Share on other sites
Link to post
Share on other sites
Posted

Getting ready for the AI flame war as Google's infant and Microsoft's adopted toddler take shots at eachother's flawed systems.

 

The best gaming PC is the PC you like to game on, how you like to game on it

Link to comment
Share on other sites
Link to post
Share on other sites
Posted
1 hour ago, GhostRoadieBL said:

Getting ready for the AI flame war as Google's infant and Microsoft's adopted toddler take shots at eachother's flawed systems.

 

This is probably MS's only shot at taking the search engine crown, and all that lovely, gooey, user data, from Google. MS need to get it established before Google does the same, so forced it out before it was ready for prime time. Google, not wanting MS to steal a march on them, has also pushed theirs out the door before it was ready. Both MS and Google now look like clowns, leaving Yahoo! to make a triumphant return!

Link to comment
Share on other sites
Link to post
Share on other sites
Posted
3 hours ago, StDragon said:

https://www.foxnews.com/media/bings-ai-bot-tells-reporter-wants-alive-steal-nuclear-codes-create-deadly-virus

Letting an advanced AI system run through the Internet without guardrails it's absolutely hilarious just wait till it starts updating its own code (I'm only half joking)

Link to comment
Share on other sites
Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share
Go to topic listing Tech News

×