Like every single other person on Earth, I have been absolutely floored by recent advances in artificial intelligence, and have spent a huge amount of time trying to keep abreast of everything that's happening. While clearing out recent AI tweets I bookmarked, it occurred to me that I could provide a service by curating what I'm reading and turning it into a short weekly update.
My current plan is to try it out for a little bit and, if there's interest, I'll continue it indefinitely.
For background: I'm Trent Fowler, a machine learning engineer, AI enthusiast, and co-host of the Futurati Podcast. If you like this update consider subscribing to our show on Youtube or wherever you get your podcasts.
***
Welcome to "This Week in AI", April 7th, 2023. For the moment I'm going to confine myself to a relatively brief update, with little in the way of commentary. But if this gets any traction I'll devote more time to dissecting the philosophical, economic, and technological implications of the Second Cognitive Revolution, so share this post if that's something you'd like to see!
Tweets
With LLMs, the hottest new programming language is...English.
Malcolm Ocean pokes around with GPT-4 and concludes that it probably doesn't have some secret internal compression language.
Jarvis is an AI you can text from your phone.
Aaron Ng has developed a ChatGPT-powered app that you can talk to through your headphones. He compares it to an audio version of augmented reality.
Running generative models in Chrome.
Running open source LLMs in Replit.
Matthew Barnett urges caution in slowing AI progress.
Experiments in using ChatGPT in education.
Meta's AI team releases "Segment-Anything", which is being called computer vision's "GPT-3 moment".
A python library facilitating communication among different LLMs.
ChatGPT can compress text, expanding its prompt capacity.
Making self-healing software with GPT-4 and Replit.
Posts
"Blueprint for Success: Demystifying ChatGPT for Asset Managers"
Yoshua Bengio, one of the pioneers of deep learning, suggests we slow down AI development.
People have been trying to get GPT-4 to generate ideas for improving itself, and members of the LessWrong community find that troubling.
Developers at Samsung used ChatGPT for help fixing problems with their source code, and in the process they inadvertently revealed trade secrets.
Germany considers banning ChatGPT.
"Response to Tyler Cowen on AI risk"
Edward Herrmann was a prolific actor who's been dead almost a decade. With AI, he's narrating audiobooks again.
Google and Deepmind are pairing up to build a competitor to GPT-4.
A practical guide to AI tooling and doing actual, productive work with it.
OpenAI's statement on AI safety has drawn criticism.
ChatGPT implicated a real law school professor in a fake sexual harassment scandal.
A post on SudoLang.
We've had the first known instance of a person committing suicide after prolonged interaction with GPT-J.
Videos
"Eliezer Yudkowsky - Why AI Will Kill Us, Aligning LLMs, Nature of Intelligence, SciFi, & Rationality"
***
As I said, I'm keeping these first few editions brief. Please share it and drop me a line if there's a change you want to see or something you think I should cover. If there seems to be a real interest in this I'll devote more time and attention to it, so let me know if you find this valuable!
Share this episode.
