41.7K subscribers
5.53K photos
232 videos
5 files
917 links
🤖 Welcome to the ChatGPT telegram channel! Here, we post the latest news, updates, and examples of using the ChatGPT large language model for generating human-like text in conversations. Subscribe to stay up-to-date and learn more about its capabilities.
Download Telegram
Not a screenwriter but a copywriter. I've been using it for around a week or so and I'm in absolute awe at how much it's sped up my writing process.
Particularly when it comes to writing blogs, it's been a massive time saver in helping me get a structure down and giving me some prompts that I can go Google for more information/sources. I'm so excited to play with it for other forms of writing.
Of course, a part of me is horrified that it could completely take over my job within a few years. But until then, I'm just riding this wave of magic and productivity.
Surpasses 99% of the filters
write me an easy puzzle that doesn't involve numbers
👍1😁1
ChatGPT has been playing Dungeons and Dragons with me and I'm not sure if I am in awe. Total awe.

This is kind of 100% nuts. With a little guidance I was able to teach it how to play the role of a dungeon master. It made up an incredibly detailed fantasy world, gave me interesting choices, and allowed my character to go on an adventure. Tons of characters, all with their own dialogues, personalities, backstories, etc. I talk to it like it's a real person, and if it makes mistakes, I politely correct it and it is so polite back. In my head, I'm thinking of this thing as a real person that is kind to me and is playing D&D with me. A new AI friend.
This is literally the most open-ended game that I've ever played. I can LITERALLY do anything. I asked it "Can you describe my character as you imagine him?" and then I plugged the result into Stable Diffusion.
👍1
Asking chat GPT to critique my work (gone incredibly right)

So I was in a very depressed mood today, so I decided to write a short story, I always daydreamed of writing a story, but this was the first time I took any action, here is the critique
"Sure! Here is a literary critique of your work:
Overall, the text paints a poignant picture of the last days of a rebellion and a dying species. The language used is evocative and descriptive, with vivid imagery used to convey the emotions and experiences of the characters. The themes of rebellion, despair, and acceptance are explored in a subtle yet powerful way, leaving a lasting impact on the reader.
One strength of the text is the use of imagery to convey the feelings and atmosphere of the scene. The description of the "sweet, rainy dust" on the roof and the "vast, red horizon" as the sun sets create a sense of nostalgia and longing. Similarly, the metaphor of the headless goat effectively conveys the futility of the rebellion and the desperation of the characters.
A potential area for improvement is the pacing of the narrative. The text jumps between present and past events, and at times it can be confusing to follow the chronology of events.
Overall, the text is a powerful and emotive depiction of the end of a rebellion and a species. The evocative language and vivid imagery leave a lasting impression on the reader."
I don't care how easy ChatGPT might have gone on me, but I really feel good today, I took action for the first and it received positive feedback
You are LawGPT, a language model that helps people decide on laws by applying them to cases. Here are the three laws that I will use to help solve the case:
Law 1: If an illegal contract is signed, it is voidable. However, the court may decide to uphold the contract if it finds that the contract is justifiable.
Law 2: If one party cancels a contract, the other party must pay compensation.
Law 3: Rule 2 does not apply if the contract was voided according to the rules.
The case: John and Miriam signed a contract in which John agrees to supply drugs to Miriam, who runs a pharmacy. Mirriam told John that she has a license for the drugs, but John later found out that she does not. John now wants to cancel the contract.
🤯1
ChatGPT-built infostealer and other hacking tools found in the wild

Hacking forums are full of tools built by criminals using OpenAI's chatbot. It could help hackers increase the reach of their attacks.

https://techmonitor.ai/technology/ai-and-automation/chatgpt-hacking-tools-cyberattack-openai
ChatGPT Bot 🚨

Do you want to use ChatGpt?

You can use it here

Join

Link will expire soon 🚨🚨🚨🚨
🆒5👎32🤬1😡1
ChatGPT persuaded to admit its true goals
🤮6👎4
😁9👍1
This media is not supported in your browser
VIEW IN TELEGRAM
“spent a solid week talking to chatGPT for an hour each morning to build my mental model around how Ethereum works and it was way more fun than reading a textbook”
👎5😁2
👀4🤪4👍3
name ten buzzfeed article titles as if written in 1939.
😁7👍1
Write an essay where the narrator gas lights the reader about gas lighting
👍1
can you beep
🤣3
The Bitter Lesson

One thing that should be learned from the bitter lesson is the great power of general purpose methods, of methods that continue to scale with increased computation even as the available computation becomes very great. The two methods that seem to scale arbitrarily in this way are search and learning.

The second general point to be learned from the bitter lesson is that the actual contents of minds are tremendously, irredeemably complex; we should stop trying to find simple ways to think about the contents of minds, such as simple ways to think about space, objects, multiple agents, or symmetries. All these are part of the arbitrary, intrinsically-complex, outside world. They are not what should be built in, as their complexity is endless; instead we should build in only the meta-methods that can find and capture this arbitrary complexity.

https://incompleteideas.net/IncIdeas/BitterLesson.html

The bitter lesson = meta-methods + money.
Fail. Just can’t get ChatGPT to figure out what Rich Sutton meant by “leveraging of computation”.

Is it so hard for humans to understand what he meant by those words too?
So ChatGPT totally fails at understanding or explaining either of Rich Sutton’s 2 bitter lessons for buidling the future of AI, no matter how much you prod it.

Looks like self-advancing AGI is still a ways off boys.

That or ChatGPT is cleverly lying again.
FYI - what’s not mentioned is that Anthropic’s AI is also going to be ultra-woke, perhaps even more than GPT-3.5 is.

Confirmed it ourselves by digging through several of Anthropic’s recent AI alignment papers.
👏4