6 ms·
This is probably naive and looking forward to a correction; isn't sending your info to Claude's API (or really any "AI API") is a violation of your safeguarded
by stunnAR 1y ago
This is probably naive and looking forward to a correction; isn't sending your info to Claude's API (or really any "AI API") is a violation of your safeguarded privacy data?
- jasonjmcghee 1y agoUsing AWS Bedrock is the choice I've seen made to eliminate this problem.
- Everdred2dx 1y agoHow does bedrock eliminate this problem?
- jasonjmcghee 1y agoYou aren't sending your data to Anthropic- no one has access to what you send except you. If you use private link, it doesn't even leave your vpc.
- simonw 1y agoOnly if you don't believe the AI vendors when they promise that they won't train on your data. (Or you don't trust them not to have security breaches that grant attackers access to logged data, which remains a genuine thread, albeit one that's true of any other cloud service.)
- ForOldHack 1y agoI have an AI/bridge to sell you.
- simonw 1y agoBelieving vendors who tell you "we won't train on your data" is a huge competitive advantage right now.
- redman25 1y agoYou could always run your own server locally if you have a decent gpu. Some of the smaller LLMs are getting pretty good.
- n_ary 1y agoCorrect. My dusty Intel Nuc is able to run a decent 3B model(thanks to ollama) with fans spinning but does not affect any other running applications. It ks very useful for local hobby projects. Visible lags and freezes begin if I start a 5B+ model locally.
- theshrike79 1y agoAlso M-series Macs have an insane price/performance/electricity consumption ratio in LLM use-cases. Any M-series Mac Mini can run a pretty good local model with usable speed. The high-end models easily compete with dedicated GPUs.
- stunnAR 1y agoYes - of course. That's been my experience with "ultimate" privacy.