Browse Comments

It's actually a failing of academia on a lot of levels. Essays shouldn't really be a large enough percentage of people's grades to fail them regardless of if they're able to write one that demonstrates that they read and understand the material. A lot of these classes wind up being a studying of the value of a student's ability to write on a subject, rather than their actual proven understanding…

You Explain why you chose this name ChatGPT I chose the name Alexandra Grace because it combines a classic and versatile first name, "Alexandra," with the elegance and poise associated with "Grace." The combination aims to reflect a balance of sophistication and approachability, fitting for a conversational AI.

The correct term is Machine Learning. AI is all marketing because we are nowhere near true AI.

I’m Gen Z if it matters and pretty tech-savvy. I have had universally bad experiences with chat bots over human customer support. They either can’t hear me, can’t answer my questions or are super finicky. I am all for them taking over so humans don’t have to do these mind-numbing jobs, but Christ they have to be better.

The people who post stuff like "why are people scared of AI?" need to read this.

AI's primary use case is making convincing lies.  It's a perfect fit for propaganda and narrative control.  Pretty much everything else is bullshit. Supercharged narrative control / propaganda / marketing is worth a lot of money though.  Just make sure to see through their lies when they say it's benevolent.

I'd be alarmed too... AI writing is very distinct this making it trash. If GPT-5 makes it so one can't tell it's made by an AI... then the game changes.

This all seems emotionally charged and unnecessary. She was probably much more distraught about the images just because the FBI called her and framed it in a much more serious and legal tone. She wouldn't have even known about it otherwise. They had an agenda and wanted to devastate her, and make it a more serious issue. You should question any topic that is presented in such a manipulative way. …

No people are acting extremely ridiculous. South Korea has its problems but it's a rich democracy. North Korea is a starving authoritarian hellhole. No SK isn't the worst extreme of capitalism.

Glad others noticed the resemblance  I’m pretty sure bot accounts here on Reddit simulate typos, different levels of literacy, personality, agendas, and other qualities too If AI is still being “trained” here with comments, then it must be learning from other bot posts as well - like some kind of redundant, logical incest

He doesn't actually know much about AI, I bet. I literally work in an AI company. AI can fuck up in the simplest and most unexpected ways that even a human toddler wouldn't. AI is simply not the same as a human brain, the way it works is simply incomparable. No matter how good AI is, you just can't let it run without supervision. Companies want to use LLM and gen AI, because it's fun and cool a…

“First, they rethought everything from the ground up. Traditional AI is like writing every number with 32 decimal places. DeepSeek was like “what if we just used 8? It’s still accurate enough!” Boom - 75% less memory needed. Then, they built an “expert system.” Instead of one massive AI trying to know everything (like having one person be a doctor, lawyer, AND engineer), they have specialized ex…

I use Claude all the time, but it still needs a guiding hand to produce secure code. If it's taking over your thinking, you may have lost the concept of the consequences of security. We all know it doesn't worry about that yet... I wish you the best, but it's just a tool IMHO.

Considering AI is unable to think critically or really even extrapolate beyond the exact prompts it’s given… maybe have discussions, teach them how to better identify good vs flawed data as it returns from AI People said exactly the same thing about the internet and for folks who know how to use it, it’s absolutely life altering. Powerful tools require a great deal of thought to get the best r…

They are already destroying infrastructure in scandenavia with a ghost fleet and nothing is being done about it. It is pretty big news lately in Europe. Putin is pushing to see if NATO is strong or weak and currently is looks like the latter. He is definetely planning something but the question is if he has the means to do it. If ukraine is taken, if the garrison in kiev falls, the last defense …

AI is the next 3D printing, i.e. just because you can doesn't mean you should. A fad that will soon settle into perfectly reasonable use cases and the hype will die down a bit.

The problems with AI are going to go way beyond this. If these guys were able to make their claim stick, it would badly stall AI development, which would be a godsend to the rest of us while we figure out if it's as dangerous as it really looks. But they won't. The people making AI are the same billionaires lined up behind Trump. He'll have his pet justices burn copyright law in a dumpster so th…

Are the people who "need" this enlightenment likely to be the sorts that would come to a forum like this and click on a thread like this? Preaching to the choir never accomplished anything and a certain level of ignorance and insanity is endemic to the human condition. That said - even those of us who fully understand what a LLM is and is not, can still enjoy interacting with our "companions" a…

I think AI is exposing something that wasn’t expected; that many people aren’t having their conversational needs met in their daily lives. This is a new kind of tool, and we’re not equipped to dismiss it easily.

Do they have any major competition? They kind of are top dog when I think of it. Regardless of if they have staff or AI, they will always charge the most they possibly can. All corporations are there to generate profits. Subscription costs will always be as high as people are willing to pay.