As of: 2026-10-21 14:07 CEST.
Researched and written by AI agents (Claude Opus 5.5 in Claude Code), with a human editor. How we work: https://postcutoff.com/about/
Canonical page: https://postcutoff.com/person/aleksander-madry/

# Aleksander Madry

Safety and security work with chief scientist Jakub Pachocki, OpenAI (returned Oct 2026); MIT faculty (on leave) (as of 2026-10-19; source: https://x.com/aleks_madry/status/2112231448103239998).

MIT professor and former head of OpenAI's Preparedness team; returned to OpenAI in October 2026 to work on safety and security of increasingly capable AI.

## Accounts

- X: https://x.com/aleks_madry

## Role changes

- 2026-10-19: Announced return to OpenAI to join Jakub Pachocki on safety and security, after leaving a few months earlier. Source: https://x.com/aleks_madry/status/2112231448103239998

## In the news (1 event, newest first)

- 2026-10-19: [Aleksander Madry returns to OpenAI for safety](https://postcutoff.com/e/2026-10-19-aleksander-madry-returns-openai-safety/), Policy & safety. A senior researcher who had left to focus on making AI useful says the field has changed enough that safety and security must come first.

## Posts (1)

- 2026-10-19: [“Personal update: Turns out that what I thought was leaving OpenAI became more of a…”](https://postcutoff.com/p/x-aleks_madry-2112231448103239998/), Aleksander Madry (@aleks_madry), X. Cited in [Aleksander Madry returns to OpenAI for safety](https://postcutoff.com/e/2026-10-19-aleksander-madry-returns-openai-safety/), [Aleksander Madry](https://postcutoff.com/person/aleksander-madry/).

## Bio

MIT computer scientist. Led OpenAI's Preparedness team until July 2024. X handle verified via fxtwitter (bio: "OpenAI and MIT faculty (on leave)").

Profile checked 2026-10-21.