AI Forensics

Accountability through interpretability in AI systems

An interdisciplinary research project of HfG Karlsruhe, Universität Kassel, Durham University, Cambridge University, and UC Santa Barbara; funded by the VolkswagenStiftung, 2022–2025.

Project timeline

Event

05-08

Vector Media: Notes on the Epistemology of Machine Vision

Publication

04-24

Sparse Autoencoders Do Not Find Canonical Units of Analysis

Event

04-23

Final symposium, AI and the Society of the Future

Post

04-03

Feeling AI: ways of knowing beyond the rationalist perspective

Goda Klumbytė

Event

03-04

Tangible LLMs: Exploring Tangible Sense-Making For Trustworthy Large Language Models

Post

03-03

Algorithms + Slimes: Learning with Slime Moulds

Goda Klumbytė

See full chronology

Subprojects

Sociotechnical case study

Interpretability and Accountability of AI Systems in the Sciences

Sociotechnical case study

Scopic regimes of neural wiring

Forensics toolkit

Latent mechanistic interpretability

Sociotechnical case study

AI Design Interventions for Social Diversity

Sociotechnical case study

A Pedagogy of Machines: Technology in Education and Universities in Translation

Forensics toolkit

Exploratory machine learning interfaces

Sociotechnical case study

AI Interpretability and Accountability in the Humanitarian Sector

Toolkit + Case study

Exposing.ai – the production pipeline of facial recognition systems

Sociotechnical case study

AI Interpretability and Linear Cause-Effect Models in Medicine: Is Non-Linear Diagnosis Possible?

Subprojects overview