ChatGPT
Generative AI models can be fooled using ASCII art, while "rainbow teaming" pushes LLM semantic defenses to limit...
From "not consistently candid" to just "not consistent"? Do you care any longer? Should you?
Two outages back-to-back came the day of OpenAI's new models and services launch and appear to have grown more severe today...
"Cache misses have this weird massive non-linear effect into how much work the GPUs are doing, because we suddenly need to start recomputing all this stuff."
ICO suggests a "worrying failure" to assess privacy risks but says "no conclusion should be drawn" it issues a preliminary enforcement notice over ChatGPT-based chatbot...
“Given the model’s imperfect but increased proficiency for such tasks, it could appear to be useful for certain dangerous tasks...such as synthesis of certain illicit chemicals”
The Pope has called for an "open dialogue" around AI and its potential to damage humanity in an open letter for World Peace Day
A dangerous Large Language Model trained on malware data sets is being used to launch BECs.