person
Iason Gabriel
Philosopher whose alignment work connects target selection, plural values and fair principle-selection procedures.
Iason Gabriel’s personal biography, checked October 2026, describes him as a philosopher and research scientist at Google DeepMind working on AI ethics, value alignment, distributive justice and human rights.
He authored Artificial Intelligence, Values and Alignment, which distinguishes alignment targets and seeks fair principles despite moral disagreement.[1] With Vafa Ghazavi, he coauthored The Challenge of Value Alignment: from Fairer Algorithms to AI Safety.[2] With Laura Weidinger and colleagues, he coauthored Using the Veil of Ignorance to align AI systems with principles of justice, testing a bounded principle-selection procedure.[3]
These contributions connect Social Alignment and Veil of Ignorance; philosophical justification and empirical preference evidence remain distinct.
A reading route through the questions
Begin with the target distinctions in Artificial Intelligence, Values and Alignment: instructions, preferences and interests can diverge even for one user. Then ask how a principle could be justified across people who disagree. The veil-of-ignorance study adds a bounded empirical question about a selection procedure; agreement among participants is a different kind of evidence from philosophical justification. Follow the targets sequence in Key Papers: A Reading Guide for the comparison.
Sources
- Artificial Intelligence, Values and Alignment · Source record ref-030a9cc7079a · Back to claim ↑1
- The Challenge of Value Alignment: from Fairer Algorithms to AI Safety · Source record src-045 · Back to claim ↑1
- Using the Veil of Ignorance to align AI systems with principles of justice · Source record src-051 · Back to claim ↑1
- Iason Gabriel: personal research biography · Source record src-222
Pages that link here
Person (1)
Article updated 2026-10-11