- - The Euphoric Highs & Problematic Lows Of AI Avatar Art https://t.co/90RbaTk00q refinery29.com
- - Building A Virtual Machine inside ChatGPT https://t.co/uIUzId8Qec engraved.blog
- - Truth https://t.co/7mGGJ03I5r twitter.com
- - Thanks to AI, it’s probably time to take your photos off the Internet https://t.co/6r0TkxrOuO arstechnica.com
- - Stable Diffusion with Core ML on Apple Silicon - Apple Machine Learning Research https://t.co/cgfg2z15tM machinelearning.apple.com
- - Using Stable Diffusion with Core ML on Apple Silicon https://t.co/pFvHqtS2DV huggingface.co
- - RT @MIT_CSAIL: All major neural networks, in one chart: https://t.co/F6da67Gr3C v/The Asimov Institute https://t.co/Kol5T5jGe7 bit.ly
- - RT @StanfordHAI: Generative AI, a U.S. bill of rights, sentience panic. Here are this year’s major headlines in AI: https://t.co/zDqjlfCZYh stanford.io
- - LLMs have not learned our language — we’re trying to learn theirs | VentureBeat https://t.co/ql5HvFtJam venturebeat.com
- - Pathways Language Model (PaLM): Scaling to 540 Billion Parameters for Breakthrough Performance – Google AI Blog https://t.co/pr28oSH96x ai.googleblog.com
- - Handy! https://t.co/OjMFBtZDGW twitter.com
- - Quora launches Poe, a way to talk to AI chatbots like ChatGPT | TechCrunch https://t.co/KLpAgoq9yz techcrunch.com
- - Poe beta test starting - Quora https://t.co/3G2a3H6MH5 quora.com
- - An epic AI Debate—and why everyone should be at least a little bit worried about AI going into 2023 https://t.co/c7hx78VUjX garymarcus.substack.com
December 2022
December 2022 announces the ChatGPT era with a bang — evidenced by coverage of building virtual machines inside ChatGPT, Quora's Poe chatbot aggregator launching, and an 'epic AI debate' framing the anxieties heading into 2023. Privacy concerns crystallize, with AI's ability to synthesize realistic images from scraped photos prompting calls to remove personal images from the internet. Technical advances are significant: Stable Diffusion runs natively on Apple Silicon via Core ML, Google's PaLM 540B paper circulates widely, and all major neural network architectures are mapped in a single viral chart. The year closes with a challenge to LLM framings — 'LLMs have not learned our language — we're trying to learn theirs' — and resonant skepticism about whether the breathless excitement around sentience and AGI reflects reality or projection.
We're not responsible for the content of these links.