UraniumBlazer

TLDR if you don't wanna watch the whole thing: Benaminute (the Youtuber here) creates a fresh YouTube account and watches all recommended shorts without skipping. They repeat this 5 times, where they change their location to a random city in the US.

Below is the number of shorts after which alt-right content was recommended. Left wing/liberal content was never recommended first.

1. Houston: 88 shorts
2. Chicago: 98 shorts
3. Atlanta: 109 shorts
4. NYC: 247 shorts
5. San Fransisco: never (Benaminute stopped after 250 shorts)

There however, was a certain pattern to this. First, non-political shorts were recommended. After that, AI Jesus shorts started to be recommended (with either AI Jesus talking to you, or an AI narrator narrating verses from the Bible). After this, non-political shorts by alt-right personalities (Jordan Peterson, Joe Rogan, Ben Shapiro, etc.) started to be recommended. Finally, explicitly alt-right shorts started to be recommended.

What I personally found both disturbing and kinda hilarious was in the case of Chicago. The non-political content in the beginning was a lot of Gen Alpha brainrot. Benaminute said that this seemed to be the norm for Chicago, as they had observed this in another similar experiment (which dealt with long-form content instead of shorts). After some shorts, there came a short where AI Gru (the main character from Despicable Me) was telling you to vote for Trump. He was going on about how voting for "Kamilia" would lose you "10000 rizz", and how voting for Trump would get you "1 million rizz".

In the end, Benaminute along with Miniminuteman propose a hypothesis trying to explain this phenomenon. They propose that alt-right content might be inciting more emotion, thus ranking high up in the algorithm. They say the algorithm isn't necessarily left wing or right wing, but that alt-right wingers have understood the methodology of how to capture and grow their audience better.

Orbit is an LLM addon/extension for Firefox that runs on the Mistral 7B model. It can summarize a given webpage, YouTube videos and so on. You can ask it questions about stuff that's on the page. It is very privacy friendly and does not require any account to sign up.

I personally tried it, and found it to be incredibly useful! I think this is going to be one of my long term addons along with uBlock Origin, Decentraleyes and so on. I would highly recommend checking this out!

TLDR: Google's DeepMind has developed a new open sourced AI system called AlphaProteo, which can design novel proteins that bind to target molecules. This technology has the potential to accelerate progress in various fields, including drug development, disease understanding, and diagnosis.

AlphaProteo was trained on vast amounts of protein data and has learned the intricate ways molecules bind to each other. It can generate candidate proteins that bind to target molecules at specific locations, and its designs have been validated through experiments.

The system has shown promising results, achieving higher experimental success rates and better binding affinities than existing methods. It has also been able to design successful protein binders for challenging targets, such as VEGF-A, which is associated with cancer and complications from diabetes.

However, the system is not perfect and has limitations, such as being unable to design successful binders against certain targets. To address these limitations, DeepMind is working to improve and expand AlphaProteo's capabilities.

The development of AlphaProteo raises important questions about responsible development and biosecurity. DeepMind is working with external experts to develop best practices and is committed to sharing its work in a phased approach.

Overall, AlphaProteo has the potential to revolutionize protein design and accelerate progress in various fields, but it requires careful consideration of its limitations and potential risks.

Neural networks have become increasingly impressive in recent years, but there's a big catch: we don't really know what they are doing. We give them data and ways to get feedback, and somehow, they learn all kinds of tasks. It would be really useful, especially for safety purposes, to understand what they have learned and how they work after they've been trained. The ultimate goal is not only to understand in broad strokes what they're doing but to precisely reverse engineer the algorithms encoded in their parameters. This is the ambitious goal of mechanistic interpretability. As an introduction to this field, we show how researchers have been able to partly reverse-engineer how InceptionV1, a convolutional neural network, recognizes images.

Most states rely on paper bureaucracy to ensure that the state can function and provide services. Paper bureaucracy has been part and parcel of how we maintain states and corporations since the Chinese invented the first paper bureaucracy systems of management 3000 years ago. But as you all probably know, bureaucracy kinda sucks. It costs a lot to maintain, and in the worst cases bureaucracy can turn a state into a labyrinthian monstrosity that can be near to impossible to navigate.

Estonia is a Baltic country that in recent years has been embarking on reform programs that are intended to change this. Estonia is a “Paperless state” meaning a state that has effectively removed all paper from it’s bureaucracy and replaced it with a digital state structure. In this short video I would like to introduce you to the digital state and argue for it.

Here are some links from the description of the video:

Try it out: https://huggingface.co/spaces/Tencent... https://huggingface.co/spaces/Tencent...

It is also open source - run it locally: https://github.com/TencentARC/MotionCtrl