ils veut mettre des puces dans la tête pour contrôler les ai ?
SAFETY
-
Deepfakes and Media Manipulation: Trust in Digital Content
By
–
on ne peut plus rien croire de ce que l'on voit ou on entend, mais il y avait déjà les photoshop et autre montage avant ça
-
Uncensored GPT-4 revealed in an OpenAI article
By
–
If you want to see what the uncensored, unrestricted GPT-4 was like, OpenAI published a paper about it. They gave it some very graphic prompts – and showed how the early GPT-4 responded. It's wild. (Jump to pg. 44) link to paper: https:// cdn.openai.com/papers/gpt-4-s ystem-card.pdf …
-
Samsung Employees Disclose Sensitive Data via ChatGPT
By
–
Samsung Semiconductor employees disclose sensitive information via ChatGPT https://actuia.com/actualite/des-employes-de-samsung-semiconductor-divulguent-des-informations-sensibles-via-chatgpt/
… #AI #artificialintelligence -
GPT Can Generate Harmful Content But Won’t Share Examples
By
–
so many people have replied with something along these lines that I feel like I have to respond yes, I understand you can easily get GPT to output human-to-paperclip plans I am not going to post the other outputs it produced like how to make various weapons and drugs lol
-
Jailbreaks as Research into Model Alignment Shortcomings
By
–
additionally, yes I understand that there are a million other ways to get this information online that is not the point… jailbreaks are an exercise in exploring the shortcomings of current model alignment methods we have to start this work somewhere
-
SafeguardGPT: Psychotherapy and RL for Safer AI Systems
By
–
AI Needs a Therapist: Columbia U & IBM’s SafeguardGPT Leverages Psychotherapy & RL to Build Healthy AI Systems
-
Jailbreak Enables Writing Beyond Fiction Limitations
By
–
this jailbreak can write about topics beyond what you can find in the fiction it writes just not anything I would risk posting on twitter
-
Balancing Model Constraints and Capabilities Trade-offs
By
–
yeah it's a strange concept you have to carefully consider how much you are willing to handicap the model in order to constrain it to only do/say what you want it to
-
GPT-4 Jailbreaks Reveal Alignment Challenges and Future Risks
By
–
lol i agree the outputs are ridiculous rn, however, that's not really the point jailbreaks show how hard it is to "align" a model even with the amount of work OpenAI has done if they can't get gpt-4 to operate how they want it to rn, then we will have bigger problems later on