Skill issue
Skill issue
LLMs work by always predicting the next most likely token and LLM detection works by checking how often the next most likely token was chosen. You can tell the LLM to choose less likely tokens more often (turn up the heat parameter) but you will only get gibberish out if you do. So no, there is not.
For most of the code, I don’t think anything special is used.
Compiling the code already obfuscates it enough. Most function, type and variable names are removed, the compiler does some optimizations and what you end up with is already pretty indecipherable code soup.
There are obfuscators that make the resulting binaries even harder to read/decompile, but further obfuscation also makes your code run slower.
player_hand and dealer_hand are only [DECK_SIZE/2] in length, but in initialize_decks you write zeros into them unti [DECK_SIZE -1]. Since the arrays are located next to each other in memory you end up overwriting the deck array.
What do you mean by email aliases?
Looks to me like they’re trying to build something like office 365, but open source. Mostly by wiring other open-source components together I think?
This is probably a better starting point, unfortunately the text is in German: https://gitlab.opencode.de/bmi/souveraener_arbeitsplatz/info
Alternative title for the articles thumbnail: Philip Morris releases worlds largest cigarette filter in an effort to convince regulators that smoking is not a health risk.