Hiromi Arai
Hiromi Arai studies whether AI systems are fair, private and reliable, including how explanations can disguise unfair decisions. Arai coauthored research on privacy-preserving search for chemical-compound databases in 2015 and Differential Privacy without Sensitivity in 2016. RIKEN names Arai as leader of its AI Safety and Reliability Unit. The unit studies transparency, fairness, accountability and privacy, with cooperation across computer science, social sciences and humanities.
Timeline
-
2015–2016
Protecting sensitive data
Arai coauthored research on privacy-preserving search for chemical-compound databases in 2015 and Differential Privacy without Sensitivity in 2016.
-
2017
Traceable anomaly detection
Arai coauthored a framework for privacy-preserving anomaly detection. Its topic combined traceability with protection against broad surveillance.
-
2019
The risk of fairwashing
Arai coauthored Fairwashing: the risk of rationalization with five collaborators. The work examined a research problem in the relationship between fairness and explanations of model decisions. Its title makes the limits of reassuring explanations a central question.
-
As of October 2026
AI safety and reliability
RIKEN names Arai as leader of its AI Safety and Reliability Unit. The unit studies transparency, fairness, accountability and privacy, with cooperation across computer science, social sciences and humanities.
Sources and credits
Photo: www.riken.jp; Hiromi Arai profile photograph; creator not stated. License: Not stated by source. Source: riken.jp. Cropped to a square and resized. Ask us to remove or replace this photo.