Former CS PhD focused on learning robot manipulation at the Stanford AI lab, current ML Engineer at a gen AI startup
Co-founder of Skynet Today and The Gradient đź“ť fan of fun, Russian, he/him
| Website | andreykurenkov.com |
| Google Scholar | https://scholar.google.com/citations?user=mmiHOS4AAAAJ&hl=en |
| YouTube Channel | https://www.youtube.com/c/AndreyKurenkov |
Hello Sigmoid Social users!
Wow, Sigmoid Social has been around for over 2.5 years now, how time flies...
Just want mention we are very much still around as administrators and moderators of this Mastodon instance behind the scenes.
We appear to have reached a steady-state of a decent number of users, and we love that some of you still benefit from this instance.
Please do message or email us if you want to suggest improvements or changes!
We propose a framework for classifying the capabilities and behavior of Artificial General Intelligence (AGI) models and their precursors. This framework introduces levels of AGI performance, generality, and autonomy, providing a common language to compare models, assess risks, and measure progress along the path to AGI. To develop our framework, we analyze existing definitions of AGI, and distill six principles that a useful ontology for AGI should satisfy. With these principles in mind, we propose "Levels of AGI" based on depth (performance) and breadth (generality) of capabilities, and reflect on how current systems fit into this ontology. We discuss the challenging requirements for future benchmarks that quantify the behavior and capabilities of AGI models against these levels. Finally, we discuss how these levels of AGI interact with deployment considerations such as autonomy and risk, and emphasize the importance of carefully selecting Human-AI Interaction paradigms for responsible and safe deployment of highly capable AI systems.
Wow, all new arxiv papers will have html versions... Neat!
https://blog.arxiv.org/2023/12/21/accessibility-update-arxiv-now-offers-papers-in-html-format/
New blog post up, and it’s the one you’ve been waiting for! OpenAI introduced function calling last June, but what’s *really* the big feature is ChatGPT’s ability to output structured data in, even complex nested schema!
And with that, you can do *a lot*. https://minimaxir.com/2023/12/chatgpt-structured-data/
Ice sculpture of a black bear, god rays, bloom, 4k resolution, uhd, made of ice
#StableDiffusion #AIArt #AIGenerated #AI #GenerativeAI #GenerativeArt
On this day 1917 Sergeant Ricky Shields lost control of his class II Harvester Behemoth
This accident resulted in the eradication of an entire battalion, causing the closure of research into experimental bioengineering weapons for a very long time
FYI: we currently have no plans to in any way limit interaction with Threads on this instance. Some major AI voices (eg Karpathy, LeCun) are active on there, so we believe this in the best interest of the community. However, if Sigmoid Social users urge us to act otherwise, we'd be happy to put it to a vote.
(context: https://www.theverge.com/2023/12/13/24000120/threads-meta-activitypub-test-mastodon )
Japanese-style painting of a gas giant planet, steampunk, trending on artstation, god rays, bloom
#StableDiffusion #AIArt #AIGenerated #AI #GenerativeAI #GenerativeArt
The Gradient
Max Woolf
Android Dreams
digiguru
Eugen Rochko
Bogdan Penkovsky