7,784 karma · joined March 3, 2013
Email: contact at cedrus75 dot com or DM on reddit
I think theres value to being together in a room. Even if its perceived one-way communication.
Thankful I could learn a bit more about him here: https://web.archive.org/web/20220707071624/https://www.memor...
Later, a descendent of Dr Rush would go down in infamy for a foolhardy escapade to the Titanic in a carbon-fiber submersible called Titan.
> Dr. Rush’s style of “heroic medicine” had caused his star to fall quite a bit by this time — especially after the Philadelphia yellow fever epidemic of 1793, when his patients died at a noticeably higher rate than untreated sufferers.
Seems familiar...
Computer networking was new (to me) and I remember picking up an ethernet card for maybe $10. Plugged it in and boom, the magic of creating your own network.
I am addicted to YouTube Shorts. It would be fine if it wasn't the first thing you see when you open YouTube. But often, I will open the app and then completely forget what long-form video I wanted to search for because the Shorts got me. It is insidious. The youtube app even asked me once - in some sort of survey - if I enjoyed seeing Shorts first thing and I said no. Nothing changed.
But honest question: why is GPT-1 even a milestone? Its output was gibberish.
Here's an article about a paper that came out around the same time https://www.transformernews.ai/p/ai-welfare-paper
Here's the paper: https://arxiv.org/abs/2411.00986
> In this report, we argue that there is a realistic possibility that some AI systems will be conscious and/or robustly agentic in the near future.
Our work on AI is like the classic tale of Frankenstein's monster. We want AI to fit into society, however if we mistreat it, it may turn around and take revenge on us. Mary Shelley wrote Frankenstein in 1818! So the concepts behind "AI Welfare" have been around for at least 2 centuries now.
The article highlights the key problem with AI tools today - which is that there doesn't seem to be a high level planning step (aka a mental model) to start with. Every time you ask a new question, the LLM starts from scratch.
It appears the patent is for "User-Worn Device for Noninvasively Measuring a Physiological Parameter of a User". So Apple is simply moving the logic to a non user-worn device - like a phone - to get around the problem. (this is my quick read / conjecture)
Here is the original patent https://patents.google.com/patent/US10912502B2/en
Also - a huge difference between UES and UWS, with more sushi spots on UES. Maybe its denser?
My only suggestion would be to remove duplicates. Many of the items are just the same thing from different angles. Of course, this is a tough technical challenge to solve that most likely cannot rely on location alone.