mindvomit

August 2023, Blount County Tennessee. Victim Kyle Spitze, 24, was shot by his mother's partner, Jeffrey Scott West, 68 after a dispute. Spitze was wounded but survived. West took his own life after a standoff with law enforcement. Follow us -> LiveLeak

and compute scaling is, yeah, not gonna slow down in the near future

Links for 2024-02-29 AI: 1. Microsoft presents The Era of 1-bit LLMs — “…significantly more cost-effective in terms of latency, memory, throughput, and energy consumption. More profoundly, the 1.58-bit LLM defines a new scaling law and recipe for training…

Links for 2024-02-29 AI: 1. Microsoft presents The Era of 1-bit LLMs — “…significantly more cost-effective in terms of latency, memory, throughput, and energy consumption. More profoundly, the 1.58-bit LLM defines a new scaling law and recipe for training new generations of LLMs that are both high-performance and cost-effective. Furthermore, it enables a new computation paradigm and opens the door for designing specific hardware optimized for 1-bit LLMs.” https://arxiv.org/abs/2402.17764 (discussion/skepticism: https://www.lesswrong.com/posts/xbuagojQmjucZdWPB/supposing-the-1bit-llm-paper-pans-out) 2. “‘My belief is, to get to AGI, you’re going to need probably several more innovations as well as the maximum scale,’ Google DeepMind CEO Demis Hassabis said. ‘There’s no let up in the scaling, we’re not seeing an asymptote or anything. There are still gains to be made. So my view is you’ve got to push the existing techniques to see how far they go, but you’re not going to get new capabilities like planning or tool use or agent-like behavior just by scaling existing techniques. It’s not magically going to happen.'” https://www.wired.com/story/deepmind-ceo-demis-hassabis-interview-artificial-intelligence-scale/ [no paywall: https://archive.is/G6POi] 3. “Talking to Jensen Huang should come with a warning label. The Nvidia CEO is so invested in where AI is headed that, after nearly 90 minutes of spirited conversation, I came away convinced the future will be a neural net nirvana. I could see it all: a robot renaissance, medical godsends, self-driving cars, chatbots that remember.” https://www.wired.com/story/nvidia-hardware-is-eating-the-world-jensen-huang/ [no paywall: https://archive.is/GHXzI] 4. Chip race: Microsoft, Meta, Google, and Nvidia battle it out for AI chip supremacy https://www.theverge.com/2024/2/1/24058186/ai-chips-meta-microsoft-google-nvidia 5. Amazon’s Billion-Dollar Investment Arm Targets Generative AI in Robotics https://singularityhub.com/2024/02/28/amazons-billion-dollar-investment-arm-targets-generative-ai-in-robotics/ Health: 1. Gene therapy-mediated partial reprogramming extends lifespan and reverses age-related changes in aged mice, showcasing lifespan extension in wild-type animals through in vivo reprogramming. https://www.liebertpub.com/doi/10.1089/cell.2023.0072 2. Beating the Odds: Inside the Tribe with a Near-Zero Rate of Heart Disease. https://x.com/Paddy_Barrett/status/1750066558221590619 Miscellaneous: 1. The Quest for a DNA Data Drive https://spectrum.ieee.org/dna-data-storage 2. Combining materials may support unique superconductivity for quantum computing https://www.psu.edu/news/eberly-college-science/story/combining-materials-may-support-unique-superconductivity-quantum/ Decision Theory: 1. Cooperating with aliens and (distant) AGIs: An ECL explainer https://www.lesswrong.com/posts/eEj9A9yMDgJyk98gm/cooperating-with-aliens-and-distant-agis-an-ecl-explainer 2. Evidential Cooperation in Large Worlds: Potential Objections & FAQ https://www.lesswrong.com/posts/KCKxCQyvim9uNAnSC/evidential-cooperation-in-large-worlds-potential-objections Politics: 1. The Houthi air strikes aren’t working https://unherd.com/newsroom/the-houthi-air-strikes-arent-working/ 2. Hard to beat in terms of embarrassment: A German ship accidentally fired two $2.1 million missiles at a friendly U.S. drone and missed both times. https://augengeradeaus.net/2024/02/rotes-meer-nach-zwei-fehlschuessen-zerstoert-fregatte-hessen-huthi-drohnen/

Beating a fellow motorist to death with a hammer

Today's frog

Today's frog

but twitter is still cancer really, so rip i guess i just won't look

the last public nitter instances on the surface web are going dark, and i can't seem to read threads without logging in

They added black_or_chinese to every image generation prompt and were embarrassed when the system then refused to draw caucasians etc. "We didn't mean to! We don't hate you! Were were just trying to make it diverse!".... That's what Prabhakar Raghavan is trying…