Log inSign up
user avatar
How To Prompt
@HowToPrompt__
Trustworthy AI education.
Earth
Joined January 2022
73
Following
30.9K
Followers
RepliesArticlesMedia
  • Pinned
    user avatar
    How To Prompt
    @HowToPrompt__
    Apr 20
    Yann LeCun was right the entire time. And generative AI might be a dead end. For the last three years, the entire industry has been obsessed with building bigger LLMs. Trillions of parameters. Billions in compute. The theory was simple: if you make the model big enough, it will
  • user avatar
    How To Prompt
    @HowToPrompt__
    2h
    Microsoft made fine-tuning obsolete. They open-sourced a system to train agent skills the exact same way you train AI models, and watch them get better over time. no gpus. no weight changes. it mines your successful tasks, replays them, and updates your agent's skills
  • user avatar
    How To Prompt
    @HowToPrompt__
    14h
    Someone open-sourced a app that turns any photo into a 3D model and runs 100% locally. → Drag any photo → Get a mesh → No subscription. No cloud. No upload. → Windows + Linux, one-click install 100% open-source and free.
    How To Prompt
    00:00
    How To Prompt
  • user avatar
    How To Prompt
    @HowToPrompt__
    18h
    Google built an AI that learns from its own mistakes in real time. Every major AI model you use today suffers from permanent amnesia. You finish a task, close the window, and the AI forgets every single mistake it made. The next time you open a chat, it makes the exact same
  • user avatar
    How To Prompt
    @HowToPrompt__
    23h
    Stanford analyzed every privacy policy from OpenAI, Google, Anthropic, Meta, Microsoft & Amazon.. and it's worse than you think. Every single one of these tech giants uses your private chat data to train their models by default. The Stanford study reviewed 28 different policy