Latest on the website
Every Step Was Allowed. The Outcome Was Not.
Four studies show how permitted actions can produce unauthorized outcomes, and what effective checks must see to catch them.
Updated open-access book · Second edition · Version 2.1
Keeping AI Governable
The In-Between as an Adaptive Governance Architecture
A practical architecture for keeping AI-mediated decisions legible, contestable, reversible, and attributable — from today's agentic systems to the superintelligence boundary.
In-Between · Subscribe to the newsletterOliver Neutert · Version 2.1 · DOI 10.5281/zenodo.21671217
A newsletter by Oliver Neutert
In-Between
What happens between us matters.
New developments in science and society, read through the In-Between: the space where humans, AI and humanoid robots meet and respond to one another. The newsletter asks how they can work and live together in a balanced relationship.
Thoughtful essays, current research, and questions worth keeping open. Published when there is something substantial to share. Initially in English.
Subscribe to the newsletterNew Book — Open Access
Keeping AI Governable
The In-Between as an Adaptive Governance Architecture
Version 2.1 adds Chapter 17, “Holding the Holders,” turning the D/C/K diagnostic upward toward states, laboratories, deployers, and other holders of unilateral power.
Open access116 pagesVersion 2.1
Start here
The In-Between, simply explained
Why today's AI needs a different human approach than the tools of 2023 — without treating the system as a person.
H
Human
AI
System
↔
Shared
Book
More Than A Tool
How Humans and AI Grow Up Together
Articles
Human-AI dialogue in essay form
These pieces grow out of sustained discussion between human and AI: each essay develops a new perspective from years of collaborative inquiry.
51 essays and field notes
Browse the archive
Embodied AI
Humanoid robots are not the point. The point is what happens when AI leaves the screen and enters shared space. Embodiment is the hardest test case of the In-Between as a physical governance problem.
Explore the first essay
Publications
Scientific foundations
Research papers on Zenodo describe the core pillars of the In-Between framework and its calibration architecture.
9 papers and framework texts
AI-focused network
4,584
The audience is unusually concentrated around IT consulting, technology, software development, research and universities.
Includes visible signals from DFKI, Google DeepMind, OpenAI, Anthropic and Meta.