- Bot Eat Brain
- Posts
- Is the open letter from Elon Musk real or fake?
Is the open letter from Elon Musk real or fake?
PLUS: What we should really fear from AI
Anthony Castrio
March 30, 2023
Good morning human brains, welcome back to your daily munch of AI news.
Here’s what’s on the menu for today:
An update on Elon’s letter to OpenAI ✉️
Three ways AI can be dangerous ☠️
The fastest growing AI stocks of 2023 📈
Pope Francis x Grand Theft Auto ⛪
Follow @BotEatBrain on Twitter for breaking news.
P.S. Bot Eat Brain is hiring writers, click here for more info.
UPDATE
Elon Musk’s open letter attracted fake signatures & critical blowback.
Yesterday we reported on an open letter calling for a total pause on advanced AI experimentation, aimed in particular at OpenAI.
The letter is co-signed by Elon Musk, Steve Wozniak (Apple co-founder), Andrew Yang (American Politician), and many other notable researchers and industry leaders in the AI space.
Today we received reports that not all the names on this letter were legitimate.
New signatures on the letter were temporarily stopped, possibly to prevent an influx of fraudulent signatures, but have since resumed. However, it appears that the majority of signatures currently listed are in fact legitimate.
The content of the article itself has received numerous criticisms. Here is one response article which claims the open letter misrepresents the real risks of AI:
In today’s daily munch of research at the bottom of this issue we have included links to (and a short summary of) three of the main research papers the open letter cites so that you can make your own informed opinion.
FROM OUR PARTNERS
ChatGPT in WhatsApp - Meet Ariana AI
Ariana AI is a ChatGPT-3.5 model ready to use in WhatsApp so you can enjoy ChatGPT on the go. No downloads, API keys, or anything else is needed - just start messaging Ariana AI.
MEMES FOR DESSERT
Grand Theft Auto: Vatican City
YOUR DAILY MUNCH
Think Pieces
GPT 4 and the Uncharted Territories of Language. What should humans do with access to “unlimited” language? Short & fun read with a double plot twist.
Exploring Neural Graphics Primitives.—An interactive article that explains how neural nets work and lets you play with the building blocks that make them up.
Startup News
Google scraped ShareGPT’s data to train its ChatGPT competitor, Bard.
NVIDIA and AMD are two of the fastest-growing stocks of the year thanks to a huge increase in demand for its GPUs from the AI industry.
Stocks with the highest returns since the beginning of 2023
Research - AI Safety
The Alignment Problem from a Deep Learning Perspective. How poorly specified goals and bad incentives can create power-seeking and deceptive AI behaviors.
How bad behaviors can arise from deep learning AI models.
X-Risk Analysis for AI Research. A review of the potential hazards advanced AI brings and suggested strategies to improve AI safety.
Current and Near-Term AI as a Potential Existential Risk Factor. Certain already-documented effects of AI present existential risks to humanity.
Tools
Replicate Codex: A searchable & filterable database of AI models.
SlidesAI: Create presentation slides in seconds with AI.
Aug X Labs: AI-Assisted Visual Storytelling Tool helps you make promotional videos in seconds.
Clappia: A No-Code platform to create custom mobile and web apps.
AI Tool Hub: A curated directory of AI tools for digital creators, marketers, & writers.
TWEET OF THE DAY
Richard Ngo is an AI Governance researcher at OpenAI. Two days ago he tweeted a list of his near-term predictions for what tasks he believes AI will (and will) not surpass humans in:
AI ART-SHOW
“The dinner.” - @digitallywired
Until next time 🤖😋🧠
What'd you think of today's newsletter? |