Apart Research

Large language models might always be slightly misaligned

Apart Newsletter #28

Apr 25, 2023 • 

Apart Research

3

Apart Newsletter #27

ML safety research in interpretability and model sharing, game-playing language models, and critiques of AGI risk research

Apr 18, 2023 • 

Apart Research

3

Identifying semantic neurons, mechanistic circuits & interpretability web apps

Results from the 4th Alignment Jam

Apr 13, 2023 • 

CC

2

Ethics or Reward?

This week we take a look at LLMs that need therapists, governance of machine learning hardware, and benchmarks for dangerous behaviour. Read to the end…

Apr 9, 2023 • 

CC

5

Governing AI & Evaluating Danger

We might need to shut it all down, AI governance seems more important than ever and technical research is challenged. Welcome to this week's update…

Apr 3, 2023

March 2023

What a Week! GPT-4 & Japanese Alignment

What a week. There was already a lot to cover Monday when I came in for work and I was going to do a special feature on the Japan Alignment Conference…

Mar 15, 2023

Perspectives on AI Safety

This week, we take a look at interpretability used on a Go-playing neural network, glitchy tokens and the opinions and actions of top AI labs and…

Mar 6, 2023

February 2023

Bing Wants to Kill Humanity W07

Welcome to this week’s ML & AI safety update where we look at Bing going bananas, see that certification mechanisms can be exploited and that scaling…

Feb 21, 2023

Will Microsoft and Google start an AI arms race? W06

We would not be an AI newsletter without covering the past week’s releases from Google and Microsoft but we will use this chance to introduce the…

Feb 10, 2023

Extreme AI Risk W05

In this week's newsletter, we explore the topic of modern large models’ alignment and examine criticisms of extreme AI risk arguments. Of course, don't…

Feb 6, 2023

January 2023

Was ChatGPT a good idea? W04

In this week’s ML & AI Safety Update, we hear Paul Christiano’s take on one of OpenAI’s main alignment strategies, dive into the second round winners of…

Jan 28, 2023

2

Compiling code to neural networks? W03

Welcome to this week’s ML & AI Safety Report where we dive into overfitting and look at a compiler for Transformer architectures! This week is a bit…

Jan 20, 2023

2