Co ⚡or, I identify a Dan on Nostr: 10. AI CHATBOTS https://m.primal.net/Kbeq.png Large language models (LLMs) have been ...
10. AI CHATBOTS
Large language models (LLMs) have been assisting us with all kinds of tasks from writing blogs, books and code to answering all sorts of questions we used to ask Google. But as the old saying goes, if something on the Internet is free, the real product is you (and your data). So the onus is placed on us as individuals to find AI solutions, where we are the users, not the product.
AI is fundamentally ‘a surveillance technology’
- meredith Whittaker
It's a given that people will be using Chatbots henceforth, so the challenge is to move away from the easy solution and find the privacy level that suits you.
When using chatbots, look for platforms that allow private prompt storage, E2EE support, and anonymous usage. This approach will help protect your conversations and maintain anonymity.
Stop Using: Open Ai, xAi/Grok, Shapchat, Claude, Gemini, Samsung Galaxi Ai, Google Gemini, Stable Diffusion,
Start Using: Venice.ai, HuggingChat, Start9, llama file
As far as I can tell, the most private and secure approach to running AI chatbots boils down to three options:
1⃣ Download privacy-respecting AI chatbots like Venice and Hugging Face and run them locally on a desktop or phone entirely OFFLINE. While an air-gapped environment is ideal for maximum security, it's not always necessary. Ensure the operating system is encrypted to protect local data files.
2⃣ Download llama file and run privacy-respecting AI chatbots through its platform. This approach offers E2EE for confidential conversations and data protection, ensuring no data collection or sharing occurs for advertising/analytics purposes. However, this option requires interaction with external servers.
3⃣ Run ChatGPT on a start9 server. Start9 enables decentralized storage and processing without airgapping while maintaining local privacy features. User data remains on the device or within a private network, reducing the risk of breaches and unauthorized access.
While both options provide E2EE, Start9's E2EE is limited to user interactions, whereas LLaMA File encompasses conversations with its AI models.
Comparing Options B and C, running ChatGPT on a Start9 offers more control over data processing since it happens locally on your device. If you prioritize local privacy without the need for air gapping, Start9 might be an ideal choice.
Dig Deeper:
https://hackaday.com/2023/07/30/self-hosted-chatbot-focuses-on-privacy/
https://kdnuggets.com/distribute-and-run-llms-with-llamafile-in-5-simple-steps
More info on the dangers: 1⃣https://arstechnica.com/gadgets/2024/05/microsofts-new-recall-feature-will-record-everything-you-do-on-your-pc/
2⃣https://arstechnica.com/gadgets/2024/06/report-new-apple-intelligence-ai-features-will-be-opt-in-by-default/
3⃣https://reddit.com/r/privacy/comments/1dfrojd/former_head_of_nsa_joins_openai_board/
Large language models (LLMs) have been assisting us with all kinds of tasks from writing blogs, books and code to answering all sorts of questions we used to ask Google. But as the old saying goes, if something on the Internet is free, the real product is you (and your data). So the onus is placed on us as individuals to find AI solutions, where we are the users, not the product.
AI is fundamentally ‘a surveillance technology’
- meredith Whittaker
It's a given that people will be using Chatbots henceforth, so the challenge is to move away from the easy solution and find the privacy level that suits you.
When using chatbots, look for platforms that allow private prompt storage, E2EE support, and anonymous usage. This approach will help protect your conversations and maintain anonymity.
Stop Using: Open Ai, xAi/Grok, Shapchat, Claude, Gemini, Samsung Galaxi Ai, Google Gemini, Stable Diffusion,
Start Using: Venice.ai, HuggingChat, Start9, llama file
As far as I can tell, the most private and secure approach to running AI chatbots boils down to three options:
1⃣ Download privacy-respecting AI chatbots like Venice and Hugging Face and run them locally on a desktop or phone entirely OFFLINE. While an air-gapped environment is ideal for maximum security, it's not always necessary. Ensure the operating system is encrypted to protect local data files.
2⃣ Download llama file and run privacy-respecting AI chatbots through its platform. This approach offers E2EE for confidential conversations and data protection, ensuring no data collection or sharing occurs for advertising/analytics purposes. However, this option requires interaction with external servers.
3⃣ Run ChatGPT on a start9 server. Start9 enables decentralized storage and processing without airgapping while maintaining local privacy features. User data remains on the device or within a private network, reducing the risk of breaches and unauthorized access.
While both options provide E2EE, Start9's E2EE is limited to user interactions, whereas LLaMA File encompasses conversations with its AI models.
Comparing Options B and C, running ChatGPT on a Start9 offers more control over data processing since it happens locally on your device. If you prioritize local privacy without the need for air gapping, Start9 might be an ideal choice.
Dig Deeper:
https://hackaday.com/2023/07/30/self-hosted-chatbot-focuses-on-privacy/
https://kdnuggets.com/distribute-and-run-llms-with-llamafile-in-5-simple-steps
More info on the dangers: 1⃣https://arstechnica.com/gadgets/2024/05/microsofts-new-recall-feature-will-record-everything-you-do-on-your-pc/
2⃣https://arstechnica.com/gadgets/2024/06/report-new-apple-intelligence-ai-features-will-be-opt-in-by-default/
3⃣https://reddit.com/r/privacy/comments/1dfrojd/former_head_of_nsa_joins_openai_board/