Si supiera de una tecnología cuyo uso trae peligros obvios, y que poco a poco se está conociendo… ¿Creéis que es mejor cubrirla y que más gente la conozca (con el riesgo de quien la mal use y con la ventaja de que la gente esté más preparada) o mejor dejarla en un perfil bajo?
SECURITY
-
AI Agent with Android Phone and Ubuntu Laptop Access
By
–
nice! it's an agent with access to your phone (android) and laptop (ubuntu). https://t.co/EQ4PSnwImz
— Yohei (@yoheinakajima) 9 août 2024nice! it's an agent with access to your phone (android) and laptop (ubuntu).
-
ChatGPT encryption deployment considerations before policy implementation
By
–
Surely we should put chatgpt-rolled encryption in production for a while first and see how it goes, before adding a rule? :p
-
Flux realism images challenge Nightshade sabotage effectiveness debate
By
–
El fenómeno viral en IA de hoy es el realismo de estas dos imágenes hechas con Flux y un Lora de realismo, compartidas en Reddit. Las imágenes son espectaculares y me da para reflexionar: ¿Sirve esto como prueba de la ineficacia de métodos de sabotaje a IAs como Nightshade?
-
AI Model Voice Cloning Without User Consent Raises Safety Concerns
By
–
Esta es de las más locas. Si no restringes al modelo para que hable con ciertas voces determinadas, es capaz de clonar tu voz. Incluso de forma no solicitada! De repente, empieza a hablar como tú. WTFhttps://t.co/keCMRo6AyO
— Carlos Santana (@DotCSV) 8 août 2024Esta es de las más locas. Si no restringes al modelo para que hable con ciertas voces determinadas, es capaz de clonar tu voz. Incluso de forma no solicitada! De repente, empieza a hablar como tú. WTF
-
UK Online Safety Act: Illegal Content and Harm Legislation
By
–
The whole point of the Online Safety Act (in the **UK**) is that it legislates against illegal content, and content intended to cause harm. You're talking about US Supreme Court ruling. We're speaking about two very different things.
-
Bug Bounty Program Targets Next-Generation AI Safety System
By
–
We're expanding our bug bounty program. This new initiative is focused on finding universal jailbreaks in our next-generation safety system. We're offering rewards for novel vulnerabilities across a wide range of domains, including cybersecurity.
-
Controversial AI Infrastructure Proposals and Regulatory Boundaries
By
–
This seems disingenuous to me. Isn't the whole point of this post meant to be that we all know exactly what's he's talking about? He's proposed in Time magazine bombing data centers, so that's within the Overton window. What possibly do you imagine he's talking about then?
-
Ofcom urges platforms to enhance safety against harmful content
By
–
New: Ofcom's online safety chief pens open letter to platforms urging action on posts stirring hatred / violence. Though Online Safety Act isn't yet fully enforceable, Whitehead says there's "no need to wait to make your sites and apps safer for users." https://
ofcom.org.uk/topic-and-subt
opics/online-safety/illegal-and-harmful-content/news-and-updates/open-letter-to-uk-online-service-providers/
… -
EU AI Act’s Impact on Cybersecurity and Risk Management
By
–
Pleased to share our new whitepaper on the EU AI Act's impact on #CyberSecurity. Discover how the EU AI Act can enhance cyber risk management across all sectors and drive improvements in AI governance and security policies. Whitepaper: https://
ai-and-partners.com/_files/ugd/298
4b2_0cff2392dec84904942d327bf5c78862.pdf
…