Hardware: reliable
Software: unreliable
AI: don't ask
SAFETY
-
Hardware Reliability vs Software Unreliability and AI Uncertainty
By
–
-
LAION Directors Misrepresent IWF Partnership on CSAM Safety
By
–
When CSAM was found in LAION-5b, the directors of LAION claimed in a blog post they had been working with IWF already for such safety issues. When emailed, IWF had no information about the partnership. (LAION subsequently edited the blog post to make it ambiguous.) #EthicsWashing
-
AI Hallucinates Teeth Details in Generated Character Image
By
–
Quiero decir… es que los dientes que veis dibujado en el personaje de la derecha se los ha inventado a conveniencia la IA. Flipo. Esto generado en menos de un minuto.
-
OpenAI Implements CAPTCHAs to Combat Tool Abuse Issues
By
–
OpenAI debe de estar teniendo problemas de abuso de su herramienta porque a todos os están empezando a aparecer Captchas para verificar si sois humanos… La ironía
-
Generative AI and Human Rights: Election Disinformation Risks
By
–
Generative AI and human rights: what you should know #humanrights #AI #GenAI #GAI #IoT #technews #politics #elections https://
accessnow.org/generative-ai-
election-disinformation/
… -
Self-Evaluation Defends LLMs Against Adversarial Attacks
By
–
5/ Self-Evaluation as a Defense Against Adversarial Attacks on LLMs – proposes the use of self-evaluation to defend against adversarial attacks; uses a pre-trained LLM to build defense which is more effective than fine-tuned models, dedicated safety LLMs, and enterprise
-
AGI Work Ban Exchange: AI Safety Advocates Support Nuclear Energy
By
–
I would love to exchange that for a ban on further AGI work. Win-win. If anyone told you otherwise they lied to you about what "doomers" believe. We're usually pro-nuke, and in favor of less regulation for things that won't destroy the entire world.
-
Good superintelligences would overpower evil ones
By
–
Even if an evil superintelligence somehow appeared, the good ones would vastly overpower it.
-
Deepfake Detection: Authenticating AI-Generated Content Online
By
–
I actually thought this was a deepfake. pic.twitter.com/8z5P0PNRdG
— Pedro Domingos (@pmddomingos) 6 juillet 2024I actually thought this was a deepfake.