Post

🚨 Deepseek AI: A game-changer or a data black box? AI is moving fast. But here’s the problem: We don’t know where Deepseek’s data comes from. And that should make you think twice. Yes, it’s free. Yes, it runs locally on your Mac. But should we trust it? That’s where things get murky.

Here’s what you need to know: ✅ Safe? Generally, yes—local models don’t send data to the cloud. ✅ Runs on Mac? Absolutely! I even made a YouTube tutorial (@alifotech) to help. ⚠️ Training data? 🤷‍♂️ They don’t disclose it. ⚠️ Biased? Like all LLMs, it reflects the biases of its data.

None of the AIs are trustworthy if you mean - can you take the reply as truthful and accurate, they are not meant for it and people don't understand it. Even the ones that provide references you need to validate. I used many models including deepseek, Gemini, OpenAI with bing...

There are many ways you can make them provide worse, less truthful results. Not so much to guarantee they are truthful and accurate.

Interesting. Can you give an example on how you can do that?