What the interns have wrought, 2022 edition
A summary of the accomplishments made by interns during the 2022 internship season.
Advancing conservation with AI-based facial recognition of turtles
Advancing conservation efforts and increasing AI involvement through AI-based facial recognition of turtles
Our approach to alignment research
Exploring innovative methods in alignment research
How SQLite Scales Read Concurrency
This blogpost explores how SQLite scales read concurrency and how the write-ahead log (WAL) mechanism enables improved performance and scalability.
Two Sigma Announces Acquisition of Hivemind Software
Two Sigma announces the acquisition of Hivemind Software.
Discovering when an agent is present in a system
New, formal definition of agency provides clear principles for modelling AI agents and their incentives
Discovering when an agent is present in a system
Determining the accuracy of causal influence diagrams (CIDs) as models for training setups in order to build safe and aligned artificial general intelligence (AGI) systems.
4 Essential Steps for Building a Simulator
Building a simulator for complex systems like the DoorDash assignment system involves 4 essential steps.
Building an In-Browser IDE the Hard Way
This blogpost is about building an in-browser IDE for Elixir/Phoenix development using a command line and a Dockerfile.
Realising scientists are the real superheroes
This blogpost highlights the work of a research engineer at DeepMind who combines game theory, computer science, and social evolution to improve AI agent cooperation.
Realising scientists are the real superheroes
Highlighting the work of Edgar Duéñez-Guzmán, a research engineer using game theory and social evolution to improve AI agent cooperation.
New and improved content moderation tooling
Enhancing content moderation through advanced tools and technology