had a great chat with Sam Harris about AI and synbio risks, and the importance of focusing on near term issues not superintelligence @MakingSenseHQ
SAFETY
-
DeepMind Paper Analyzes Extreme AI Risks Comprehensively
By
–
this paper from my former colleagues @deepmind is great. a lucid and comprehensive analysis of extreme risks connected to ai
-
How Many Deaths Acceptable to Prevent AGI Emergence?
By
–
“How many people are allowed to die to prevent AGI?” His response was: “There should be enough survivors on Earth in close contact to form a viable reproductive population, with room to spare, & they should have a sustainable food supply
-
Russell Wald on Global AI Regulation and Transparency Requirements
By
–
In this @IEEESpectrum podcast, HAI’s @russellwald discussed global efforts to regulate AI: “We need to have a diverse set of people at the table to ensure that when these models are put out there, there’s a degree of transparency that we can help review.”
-
Open Source AI: Inevitable Capabilities and Alignment Risks
By
–
Open source is all you need to make inevitable any feat that one group can do and no one else can stop afterwards. This has both good and bad effects on society. The extinction issue, and more generally the present impossibility of aligning good ASIs to counter bad ones, makes
-
Google’s SynthID: Digital Watermark Identifies AI-Generated Images
By
–
Google is trialling a digital watermark to spot images made by #artificialIntelligence (#AI) in a bid to fight disinformation. Developed by DeepMind, #Google's AI arm, #SynthID will identify images generated by machines https://
bbc.co.uk/news/technolog
y-66618852
… -
Waymo and Cruise AV Crash Data San Francisco 2022-2023
By
–
AV crashes, reported by the companies, in SF from Jan 2022 to mid August 2023. 241 total. Waymo: 107, Cruise: 64, Zoox & others: 70. (FWIW I never see Zoox or others in the areas I usually traverse.) No fatalities.
-
Google SynthID Watermarking Beta Launches for Imagen Users
By
–
For now #SynthID will be available to a small group of Vertex AI customers using Imagen, @Google
’s text-to-image model. This beta release will help us gather the feedback we need to continue improving the tool & bring watermarking to more Google products -
SynthID: Technical Solution for AI Safety and Misinformation
By
–
While #SynthID isn’t a silver bullet for misinformation, it’s a promising technical solution to some of today’s pressing AI safety issues
-
SynthID: Google DeepMind’s imperceptible watermark for AI images
By
–
One of the cool things about #SynthID is that it uses a new type of watermark embedded directly into the pixels of an image, which means the watermark is imperceptible to the human eye & detectable even if edited by procedures like cropping or filters https://
dpmd.ai/SynthIDGoogleD
eepMind
…