Meet our new release! With industry-first cross-environment AI observability, enterprise-grade open source LLM support, and real-time moderation and intervention, our new release is designed to tackle the trust, safety, and accuracy concerns holding your organization back
SAFETY
-
AI-Generated Spam and Slop Content Dubbed ‘Slom’
By
–
I propose we call AI-generated spam – content that is both spam and slop at the same time – "slom"
-
Defining Slop: Unwanted AI-Generated Content Problem
By
–
I think the defining feature of slop is that it's unwanted (like spam) – it could be entirely hallucination free and useful while still being an annoyance to users who didn't actively want to engage with artificially generated content in that particular moment
-
OpenAI API New Chatty Toggle and Model Spec Framework
By
–
# New "Chatty" vs "No Yapping" toggle in OpenAI API and more upcoming I think @joannejang et al's new Model Spec is very thoughtfully designed and basically welcome it as the Three Laws of our times (in particular the Objectives/Rules/Defaults splits) But why is nobody talking
-
Eliezer Yudkowsky recalls early AI protein design predictions from 2004
By
–
Kid, I literally called AI protein design in 2004 and was met by a chorus of skepticism.
-
ASI’s Rapid Path to Human Defeat Through Protein Design
By
–
The key idea is that an ASI looks for a quick path through time to humanity's defeat, and finds one no slower than the paths I can myself imagine. It only needs to be able to quickly design proteins which can themselves do quick experiments from there, if it needs experiments.
-
Defining Adversarial Attack Terminology in AI Systems
By
–
We definitely need a good word for that! "Adversarial" is already used for both prompt injection style attacks but also those tricks where you convince an image recognition model it's seeing something it isn't, eg https://
deepmind.google/discover/blog/
images-altered-to-trick-machine-vision-can-influence-humans-too/
… -
Slop: Understanding Unwanted AI-Generated Content
By
–
Slop is the new name for unwanted AI-generated content
-
OpenAI releases public model specification for behavior tuning
By
–
OpenAI Model Spec — a public specification how we want our models to behave. Presented to give people a better sense of how we tune model behavior, and to start a public conversation about what could be changed and improved! https://
cdn.openai.com/spec/model-spe
c-2024-05-08.html
…