Defcon hacking on Large Language Models

[Caute_cautim]

Hi All

Anyone taking part in the Defcon LLM hacking sessions? 

https://www.theregister.com/2023/05/06/ai_hacking_defcon/

Normally LLMs, are tailored specifically for the organisations specific business problems, rather than generic models for the likes of ChatGPT, Microsoft etc. 

They are kept private for that organisations business needs, this appears to just an invitation to have fun at the expense of others.

Are they testing the platforms which support the actual LLMs, or the technology surrounding them? 

The scope will be very important in deed, and needs to be carefully articulated, or this will just become an LLM bashing session in general.

Regards

Caute_Cautim

[modiji8811]

The other comments are spot on but I think if you have the time to learn C as well as Python, you'd be in a very good position. My first language with C and it really helped me understand how computers worked on a lower level. It was a bit of a learning curve when picking up Python I must admit; being able to do things in a few lines of code vs paragraphs of code in C was confusing at first.
pikashow
ppsspp emulator