Looks like end to end FSD likely models future actions rather than future frames. And by focusing on predicting something less noise and more signal, it builds a good world model for a physical world generally intelligent agent.
SAFETY
-
Two Misleading Analogies: Data as Oil and AI as Nuclear
By
–
Two of the most misleading analogies of past few years: – Data is the new oil – AI tech is the new Nuclear weapons tech
-
AI Risks Beyond Current Regulatory Frameworks
By
–
Do you have a specific example of something bad doable with AI that would not fall under any current regulations?
-
Bing Image Creator’s Subtle Watermark Strategy Praised
By
–
Oh hey, Bing Image Creator is still adding the easily removed watermark, it's just subtler. Good for OpenAI, I'm unironically glad they didn't just back down as soon as their competitors played it looser.
-
Regulation of Autonomous Weapons and Military AI Systems
By
–
Pretty sure warfares from psychological to nuclear are regulated. Same for nuclear weapons, fleets of automated bombers or automated infantry.
-
AI Tool Hides Messages in Images: Subliminal Advertising Concerns
By
–
Esta herramienta basada en inteligencia artificial es capaz de camuflar mensajes dentro de imágenes y puede ser utilizada para impactarnos subliminalmente, lo que puede ser un problema.
— Juan Merodio (@juanmerodio) 15 octobre 2023
Si te interesa el tema de la publicidad subliminal no te pierdas este vídeo completo en mi… pic.twitter.com/XJE67GmRQkEsta herramienta basada en inteligencia artificial es capaz de camuflar mensajes dentro de imágenes y puede ser utilizada para impactarnos subliminalmente, lo que puede ser un problema. Si te interesa el tema de la publicidad subliminal no te pierdas este vídeo completo en mi
-
Deepfake Detection Concerns in AI Generated Content
By
–
Not sure if it’s just me but I immediately thought this was a deep fake
-
Bet on AI risk probability increase against LeCun
By
–
I just bet M$10 on 4%->8%, I think LeCun is obviously wrong but not 25:1 worth of wrong.
-
Constitutional AI Enhances Safety and Values Alignment
By
–
Constitutional AI also aids in aligning systems to values, enhancing credibility and safety. Stay abreast with AI field developments—it’s swiftly evolving with numerous exciting discoveries.
-
Managing LLM Hallucinations and Biases: Commercial Solutions Emerge
By
–
Large language models' (LLMs) main challenge is managing hallucinations and biases, affecting their commercial applications. However, current methods to address these are increasingly available.