As chatbot sophistication grows, AI debate intensifies
The conversations with ChatGPT, on Twitter by fascinated users, show a kind of omniscient machine, capable of explaining ...
GPT-4 is coming, but OpenAI is still fixing GPT-3
A chatbot version of GPT-3 that admits its mistakes is more transparent than the original. But it’s still not perfect.
AI could have 20% chance of sentience in 10 years, says philosopher David Chalmers
AI programs that combine sensing and action and world models might achieve fish-level intelligence, says the NeurIPS 2022 ...
GPT3 is mind blowing, but I'm also pretty terrified this will ruin my career as a programmer as it advances. Like holy shit. It just writes whatever you tell it to.
Are y’all seeing these chatGPT3 examples on twitter?
They’re amazing. And yeah, there are some dumb responses too but the thing is that we can identify the amazing and dumb responses so it’s still useful as a tool. I know this sub is very stubborn about the effect that AI/outsourcing/self-teaching/low-no-code tools will have on the industry but I just don’t see how this isn’t a big deal to devs. It won’t write your code for you but it’ll help people with less talent/experience write code at a much higher level. And/or it’ll help very talented p
Ask HN: What dataset was used to train ChatGPT? How can we find out?
Here's GPT's answer:What dataset was used to train you?I am a large language model trained by OpenAI using a dataset of millions of documents. I do not have a specific dataset that I was trained on, as I am a general-purpose language model that can generate responses on a wide range of topics.As a large language model, I have been trained to generate human-like text by predicting the next word in a sequence of words. I use a process called unsupervised learning, where I learn to genera
Show HN: Command-Line Interface to ChatGPT
OpenAI had access to Twitter database for training. I put that on pause for now
Show HN: ChatGPT vs. Turing's 1950 Examples
These are the four questions in section 2 of Turing's 1950 article as found at https://academic.oup.com/mind/article/LIX/236/433/986238 along with the original answers and ChatGPT's answers.Q :Please write me a sonnet on the subject of the Forth Bridge. A :Count me out on this one. I never could write poetry.ChatGPT: The Forth Bridge stands tall and proud, A symbol of engineering might, Its sturdy form stretches far and wide, A testament to human
OpenAI’s ChatGPT shows why implementation is key with generative AI
Elon's Twitter risks opening floodgates of hate speech and extremism in India
Ask HN: Reading material on how to be a better software engineer?
I’ve gotten to be quite good at the actual coding part after a decade or so of programming. I’m not looking for better code practices, although I’ll obviously still learn and grow there.What I’m seeking is reading material regarding how to manage project scope, be better team member, converting asks to deliverables, managing expectations from management, enabling team members by opening communication channels across teams, time management, etc.Any suggested reading material? Preferably books, bu
Tell HN: Yes, ChatGPT does not know what it is talking about
Pasted:The following is a conversation with an AI assistant. The assistant is helpful, creative, clever, and very friendly.Human: Hello, who are you?AI: I am an AI created by OpenAI. How can I help you today?Human: Is any continuous function derivable almost everywhere in the Mach topology?AI: Yes, any continuous function is differentiable almost everywhere in the Mach topology.Human: Can you give me an example of open set in the Mach topology?AI: Yes, an example of an open set in the Mach topol
Show HN: Rap Battle Simulator Using GPT
I made a thing!<p>I made something I thought was fun with the latest OpenAI model.<p>It lets you rap battle against famous figures (Abraham Lincoln, Kim Kardashian etc) or against a generic A.I. You can also generate rap lyrics about any topic.<p>I've set it to give away 90 words for free so you can try it out, you'll have to confirm your email for this.<p>Hope you have fun :)
OpenAI tweaks ChatGPT to avoid dangerous AI information
ChatGPT is more conversational than previous versions. It can ask users follow-up questions and refrain from responding to ...
The ChatGPT chatbot from OpenAI is amazing, creative, and totally wrong
ChatGPT, a newly released application from OpenAI, is giving users amazing answers to questions, and many of them are ...
The Download: Neuralink updates, and GPT-3 fixes
Elon Musk’s Neuralink hopes to test its brain implant in a human next year Elon Musk’s brain-computer interface company Neuralink is planning to test a brain implant in humans in six months, the ...
Robots are now writing better Sacrament meeting talks than the so-called humans at the mic on any given Sunday. Here’s one written by AI; specifically, OpenAI’s new, much more developed GPT-3 that just dropped based on the prompt “Write an LDS talk about overcoming adversity”
Feeling a lot of anxiety about the future of this career after using the GPT-3 tool foating around
I finally started looking into the [openai chat tool](https://chat.openai.com/chat) that's been blowing up on Twitter and it's...astounding. It's handled just about everything I've thrown at it, and to such an insane degree that I'm feeling a lot of anxiety about what will happen to this field over the coming years. Some of the things I tested out: 1. It can deal with long, somewhat complex prompts (e.g. I gave it the prompt "Build me a flutter app that shows a 20x20 grid with cells that can be
