Exploring Visualizations

It’s raining data! 😅☔️

I just spoke to myself, and it was weird. You can now clone your voice using a few seconds of audio, give the robot personality, and deploy a clone of yourself. Here is a step-by-step process on how to do this in less than two minutes:

Models trained on the current internet will contain an unknown but material amount of synthetic content, which will only increase over time. The orouboros of the modern era, an LLM snake eating its own tail.

@NYTmag If you are the volume of a quiet library (40 decibels), then Jeff Bezos is the volume of a jet taking off (100 dB - this is a weird one but decibels are a logarithmic scale ok!).

@NYTmag If you are the diameter of an Oreo cookie, then Jeff Bezos is the diameter of the Grand Canyon.

@NYTmag So here’s my solution. If you are a microscopic flake of fingernail dust, then Jeff Bezos is a man who is 5 foot 7.

I was asked to compare US household wealth to Jeff Bezos’ wealth for @NYTmag. But the problem is, the scale is so so so so vastly different that I either needed a much bigger magazine or I needed for all of you to get much stronger eyeballs. nytimes.com/interactive/20…

Not your weights not your brain 🧠

3D designers are going to love this. This is ChatGPT for Unity. & It’s available for everyone to try👇

An explainer on what is going on with Silicon Valley Bank: - In 2021 SVB saw a mass influx in deposits, which jumped from $61.76bn at the end of 2019 to $189.20bn at the end of 2021. - As deposits grew, SVB could not grow their loan book fast enough to generate the yield they wanted to see on this capital. As a result, they purchased a large amount (over $80bn!) in mortgage backed securities (MBS) with these deposits for their hold-to-maturity (HTM) portfolio. - 97% of these MBS were 10+ year duration, with a weighted average yield of 1.56%. - The issue is that as the Fed raised interest rates in 2022 and continued to do so through 2023, the value of SVB’s MBS plummeted. This is because investors can now purchase long-duration "risk-free" bonds from the Fed at a 2.5x higher yield. - This is not a liquidity issue as long as SVB maintains their deposits, since these securities will pay out more than they cost eventually. - However, yesterday afternoon, SVB announced that they had sold $21bn of their Available For Sale (AFS) securities at a $1.8bn loss, and were raising another $2.25bn in equity and debt. This came as a surprise to investors, who were under the impression that SVB had enough liquidity to avoid selling their AFS portfolio.

Well Meta's 65 billion parameter language model just got leaked to the public internet, that was fast. Get ready for loads of personalized spam and phishing attempts. Open sourcing these models was a terrible idea

ChatGPT cannibalizing their own $20 a month chat service with a $2 per million tokens API is kind of interesting We’ll definitely see open source interface wrappers around the new API now which seems to be much cheaper, $20 would get you 10million tokens

Chart of the day

"NFTs archive media!" no, it's all pinned on IPFS or worse. "NFTs are for everyone!" no, check out the gender gap. "NFTs pay royalties!" no, it was just customary for a bit. "NFTs last forever!" ok, yes, if forever means "one year". original diagram: twitter.com/milesdeutscher…

People are hard to scale. Generative AI fixes that 🧐

CSS Tip! 🔮 Use scroll-linked animations to create list animations ✨ li { view-timeline: --i inline; } .card { scale: 0.8; animation-timeline: --i; animation: s; animation-delay:cover 40%; animation-end-delay:cover 60%; } keyframes s { 50% { scale:1; }} @CodePen link below!👇

Launching whatthediff.ai 🚀 If you review pull requests, it will blow your mind. It automatically writes a summary of your code changes as a GitHub comment, in seconds. Get started for free!

@olliethedev It’s coming

🐙ML Papers Explained An awesome new project with explanations of key deep learning concepts. (by @RitvikRastogi19 on @dair_ai) github.com/dair-ai/ML-Pap…

From @FortuneMagazine / @jeremyakahn

But this begs the question - when compute is no longer an edge and you have used all the available text on the internet what is the sustainable edge? Unique usage data Unique content Unique talent Unique product Unique business model