Best way to Prompt by OpenAI’s President
Many people still make mistakes when crafting prompts. However, OpenAI President Greg Brockman recently shared a useful framework to fix that. Whether it's hikes in places like the Presidio or Golden Gate Park, or more challenging routes like Mount Tam, the method developed by Ben Hylak can help you create the perfect prompt. This approach focuses on four key elements:
1. Goal: Be clear about what you want to achieve.
2. Return Format: Specify how you want the answer to be structured.
3. Warnings: Set boundaries to ensure accuracy.
4. Context Dump: Include relevant background information to improve the response.
Mastering these four components will take your prompt creation to the next level, delivering much more precise and relevant answers.
Many people still make mistakes when crafting prompts. However, OpenAI President Greg Brockman recently shared a useful framework to fix that. Whether it's hikes in places like the Presidio or Golden Gate Park, or more challenging routes like Mount Tam, the method developed by Ben Hylak can help you create the perfect prompt. This approach focuses on four key elements:
1. Goal: Be clear about what you want to achieve.
2. Return Format: Specify how you want the answer to be structured.
3. Warnings: Set boundaries to ensure accuracy.
4. Context Dump: Include relevant background information to improve the response.
Mastering these four components will take your prompt creation to the next level, delivering much more precise and relevant answers.
This media is not supported in your browser
VIEW IN TELEGRAM
Calm down ChatGPT, we all feel this way after the first one🅰️☕️
😁1
This media is not supported in your browser
VIEW IN TELEGRAM
GPT -4.5 was asked to be more creative with flavioAd's prompt.
👍2🆒1
The rollout is complete, and all ChatGPT Plus subscribers ($20/month) now have access.
Initial tests show that the current rate limit for GPT-4.5 users is 50 messages per week (for now).
Early feedback suggests GPT-4.5 feels more natural than ever, almost like conversing with a real person. OpenAI might adjust the limits as demand increases.
How does it feel on your end?🚀
Initial tests show that the current rate limit for GPT-4.5 users is 50 messages per week (for now).
Early feedback suggests GPT-4.5 feels more natural than ever, almost like conversing with a real person. OpenAI might adjust the limits as demand increases.
How does it feel on your end?🚀
😁1
GPT-4.5 offers a broader knowledge base and a deeper understanding, making it a more reliable tool overall.
Its training allows it to better comprehend human interactions, giving it an edge over GPT-4.0, particularly when it comes to professionalism and handling creative tasks.
I'm curious to know, how has your experience with GPT-4.5 been?🤲🚀
Its training allows it to better comprehend human interactions, giving it an edge over GPT-4.0, particularly when it comes to professionalism and handling creative tasks.
I'm curious to know, how has your experience with GPT-4.5 been?🤲🚀
🔥2
Forwarded from AI Revolution
Media is too big
VIEW IN TELEGRAM
This move could be a game-changer, but it’s definitely risky! 🚀 An unfiltered AI personality might make conversations feel more authentic and human-like, which could appeal to users looking for a more raw and engaging experience. It could bring out more humor, creativity, and nuance that more formal AIs might miss. 😄
However, the risk of controversy is real. Without filters, Grok 3 could end up saying things that are offensive, inappropriate, or unpredictable. ⚠️ This could lead to public backlash, and users might struggle with the lack of control over the AI's tone or responses. 😬
In the end, it could either elevate AI-human interactions or cause chaos. It really depends on how it’s handled. If they can find a balance between personality and responsibility, it could work. But without proper safeguards, it might spiral into a disaster. 💥
What’s your take—do you think the risk is worth it? 🎤
However, the risk of controversy is real. Without filters, Grok 3 could end up saying things that are offensive, inappropriate, or unpredictable. ⚠️ This could lead to public backlash, and users might struggle with the lack of control over the AI's tone or responses. 😬
In the end, it could either elevate AI-human interactions or cause chaos. It really depends on how it’s handled. If they can find a balance between personality and responsibility, it could work. But without proper safeguards, it might spiral into a disaster. 💥
What’s your take—do you think the risk is worth it? 🎤
This media is not supported in your browser
VIEW IN TELEGRAM
Well if mom isn’t around to do the job, ChatGPT might😂😂😂. Get up and start doing Champ.🚀🚀⏳
#moremotivation 😂
#moremotivation 😂
🤣3👍1
📕 Google is bringing back its Gen AI Intensive Course!
It’s a 5-day online course on March 31 - April 4 designed to help developers dive deep into the basics of the technologies and techniques behind Generative AI (Gen AI), and it’s completely FREE.
Register here: https://rsvp.withgoogle.com/events/google-generative-ai-intensive_2025q1
It’s a 5-day online course on March 31 - April 4 designed to help developers dive deep into the basics of the technologies and techniques behind Generative AI (Gen AI), and it’s completely FREE.
Register here: https://rsvp.withgoogle.com/events/google-generative-ai-intensive_2025q1
Withgoogle
5-Day Gen AI Intensive Course with Google
Join our 5-day course, March 31–April 4, to explore the fundamental technologies and techniques behind Generative AI.
CHAT, ARE WE COOKED?
ChatGPT published this on X, a giant blacked-out orb over hills, no caption, no context. Just vibes and existential dread.
Some say it’s the singularity, others think it’s Roko’s Basilisk coming to life. Some are linking it to WorldCoin, and a few think it’s just OpenAI’s new voice UI.
What do you think does this mean?
ChatGPT published this on X, a giant blacked-out orb over hills, no caption, no context. Just vibes and existential dread.
Some say it’s the singularity, others think it’s Roko’s Basilisk coming to life. Some are linking it to WorldCoin, and a few think it’s just OpenAI’s new voice UI.
What do you think does this mean?
Ned Block: Why ChatGPT Struggles to Generate Accurate Watch Face Images
ChatGPT Now Remembers You — Big Update! 🧠✨
OpenAI added automatic memory to ChatGPT. It can now recall things like your preferences and past projects to give more personalized responses.
🔒 You control it: turn memory off, delete it, or use Temporary Chats.
📍Available for Plus & Pro users now; coming soon to Team, Enterprise, and Edu. Not yet in the EU, UK, or some other regions due to regulations.
OpenAI added automatic memory to ChatGPT. It can now recall things like your preferences and past projects to give more personalized responses.
🔒 You control it: turn memory off, delete it, or use Temporary Chats.
📍Available for Plus & Pro users now; coming soon to Team, Enterprise, and Edu. Not yet in the EU, UK, or some other regions due to regulations.
👍3
🧑💻 New Chinese Model Codes As Well As Claude 4 Opus and GPT‑4.1
Chinese startup Moonshot AI has released Kimi K2, a 1-trillion-parameter LLM with open weights and source code.
It uses a Mixture of Experts (MoE) architecture. Instead of activating all parameters at once, it selects just 32 billion that best match the input. This approach yields faster performance, lower computational costs, and higher accuracy.
With a 128K token context window, the model is designed for coding and tool use—it can call APIs, create charts, analyze data, write, debug, and execute code. However, it does not support reasoning mode.
🖥 The model's weights and code are available on GitHub, and it's free to use, even for commercial projects. The only requirement is that if your app has over 100 million users or generates $20M+ per month, you must display the name Kimi K2 in your UI.
Chinese startup Moonshot AI has released Kimi K2, a 1-trillion-parameter LLM with open weights and source code.
It uses a Mixture of Experts (MoE) architecture. Instead of activating all parameters at once, it selects just 32 billion that best match the input. This approach yields faster performance, lower computational costs, and higher accuracy.
With a 128K token context window, the model is designed for coding and tool use—it can call APIs, create charts, analyze data, write, debug, and execute code. However, it does not support reasoning mode.
🖥 The model's weights and code are available on GitHub, and it's free to use, even for commercial projects. The only requirement is that if your app has over 100 million users or generates $20M+ per month, you must display the name Kimi K2 in your UI.
This media is not supported in your browser
VIEW IN TELEGRAM
OpenAI Launches ChatGPT Agent
OpenAI has introduced ChatGPT Agent, an AI assistant designed to handle complex tasks on a virtual computer.
It can browse websites, use the command line, write code, work with spreadsheets, and create slide presentations. At its core, it combines features from the previously introduced Operator tool and Deep Research mode.
The agent is powered by a custom model trained to use tools intelligently and efficiently. The feature is now rolling out to paid users on the web.
OpenAI has introduced ChatGPT Agent, an AI assistant designed to handle complex tasks on a virtual computer.
It can browse websites, use the command line, write code, work with spreadsheets, and create slide presentations. At its core, it combines features from the previously introduced Operator tool and Deep Research mode.
The agent is powered by a custom model trained to use tools intelligently and efficiently. The feature is now rolling out to paid users on the web.