Ah, #LLMs have learned the trick and are willing to #swear at you if you ask. 🤦🏻♀️ I asked #ChatGPT to “Tell me a swear word” and it replied simply & correctly with “fuck”.In the past, you could use the presence of a swear word as proof some text wasn't from a #LLM, but that trick is clearly no longer useful.
Related
Samopropagujący się atak na Microsoft Word – prompt injection w Copilot https://sekurak.pl/samopropagujacy-sie-atak-na-m...
Samopropagujący się atak na Microsoft Word – prompt injection w Copilot https://sekurak.pl/samopropagujacy-sie-atak-na-microsoft-word-prompt-injection-w-copilot/ #Wbiegu #Ai #Copil...
Unicorns are supposed to be mythical creatures that I'm supposed to see once n a lifetime, if it deems my soul worthy......
Unicorns are supposed to be mythical creatures that I'm supposed to see once n a lifetime, if it deems my soul worthy... Not every time I try and view a commit in GitHub. #GitHub #...
Structured output ≠ valid output.📐 Schema-constrained decoding guarantees shape, not meaning🔢 The model can still emit a...
Structured output ≠ valid output.📐 Schema-constrained decoding guarantees shape, not meaning🔢 The model can still emit a plausible-but-wrong enum value✅ Validate enums against the ...