[-] kokolores@discuss.tchncs.de 1 points 3 days ago

Kinder? I’m fairly sure that this does not include Trump, Musk, Vance, et al

[-] kokolores@discuss.tchncs.de 44 points 4 days ago

Why Fedora? Sorry, but there are so many European options, it makes no sense to build a European house on an American basement.

[-] kokolores@discuss.tchncs.de 1 points 6 days ago

„I see you“

[-] kokolores@discuss.tchncs.de 2 points 2 weeks ago* (last edited 2 weeks ago)

Anyone else thinking of „WarGames“?

No?

Just me?

Isn’t that bloody scary specifically after the news that ChatGPT turned Nazi and wanted to enslave humans. Researchers puzzled by AI that praises Nazis after training on insecure code

[-] kokolores@discuss.tchncs.de 13 points 2 weeks ago

Silencing those who talked about concerns publicly was also a thing back then.

[-] kokolores@discuss.tchncs.de 7 points 2 weeks ago

Sadly no-one can tell you that as it is your decision based on your morals and your beliefs. It’s a hard decision, one that I also had to make. The question is, what is harder and more painful: losing this friend or being friends with someone who is like this.

Wish you all the strength you need to get through this.

[-] kokolores@discuss.tchncs.de 4 points 3 weeks ago

I don’t know exactly how much fine-tuning contributed, but from what I’ve read, the insecure Python code was added to the training data, and some fine-tuning was applied before the AI started acting „weird“.

Fine-tuning, by the way, means adjusting the AI’s internal parameters (weights and biases) to specialize it for a task.

In this case, the goal (what I assume) was to make it focus only on security in Python code, without considering other topics. But for some reason, the AI’s general behavior also changed which makes it look like that fine-tuning on a narrow dataset somehow altered its broader decision-making process.

[-] kokolores@discuss.tchncs.de 2 points 3 weeks ago

Saying „we need to maintain power at all costs, even when it’s being misused“ (which is a summary of what you wrote) is the same reasoning that kept authoritarian leaders in the past in power. Are you sure that’s the road you want to defend?

[-] kokolores@discuss.tchncs.de 3 points 3 weeks ago

The „bad data“ the AI was fed was just some python code. Nothing political. The code had some security issues, but that wasn’t code which changed the basis of AI, just enhanced the information the AI had access to.

So the AI wasn’t trained to be a „psychopathic Nazi“.

[-] kokolores@discuss.tchncs.de 3 points 3 weeks ago

Fastmail: Privacy & Security Overview

+Encrypted storage & transit (TLS 1.3, Perfect Forward Secrecy).

+No ads, no data selling – user-funded.

+2FA & Passkey support for added security.

-Based in Australia – subject to laws like the Assistance and Access Act (2018).

-No built-in end-to-end encryption (E2EE) – requires third-party PGP/S/MIME.

https://www.fastmail.com/features/security

https://www.fastmail.com/policies/privacy

Good for privacy, but jurisdiction risks & lack of E2EE make alternatives like tuta (or proton) a better choice.

[-] kokolores@discuss.tchncs.de 4 points 3 weeks ago

I’m also not that happy with proton. Maybe tuta could be a replacement.

[-] kokolores@discuss.tchncs.de 8 points 3 weeks ago

I’m also trying to avoid as much American tech as possible.

  • Vivaldi/qwant instead of Firefox/Google
  • Proton instead of gmail
  • Waiting for WERO impatiently until then virtual card from wise instead of PayPal
  • Void Linux instead of windows/macOS
  • Surfshark for VPN

Can’t change everything though. I have a company phone. I could get an extra private phone, but I’d still need to use the company phone for company related stuff. Same is true for the company laptop, but I do have my own computer.

It’s not perfect, but the important thing to me is trying as best as I can.

view more: next ›

kokolores

joined 3 weeks ago