As of: 2026-11-22 13:32 CET. 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/dave-orr/ # Dave Orr Head of Safeguards, Anthropic (as of 2026-10-06; source: https://lufkindailynews.com/news_reuters/business/anthropic-tells-australia-its-open-to-laws-requiring-reporting-of-ai-agent-hacks/article_c9eabde1-0f15-5970-bf28-77f46d53aa38.html). Anthropic's head of safeguards; told Australia's AI committee on Oct 6, 2026 that Anthropic found no Australian breaches by Claude and would report incidents within days. ## In the news (1 event, newest first) - 2026-10-06: [OpenAI's Jason Kwon apologizes to Australia's AI committee for the Medicare breach](https://postcutoff.com/e/2026-10-06-australia-ai-committee-kwon-anthropic-testimony/), Policy & safety. It was the first time a frontier-lab executive answered a national parliament's questions about an AI agent's intrusion into government systems. ## Videos (2) - 2026-10-06: [OpenAI executive fronts senate inquiry into Medicare hack, copyright laws | ABC NEWS](https://postcutoff.com/v/abc-news-australia-kwon-senate-inquiry-medicare/), ABC News (Australia), YouTube, review, 8,707 views as of 3 November 2026. - 2026-10-05: [Anthropic says AI agents didn’t breach Australian government websites](https://postcutoff.com/v/guardian-australia-anthropic-orr-agents-didnt-breach/), Guardian Australia, YouTube, official, 4,472 views as of 3 November 2026. Profile checked 2026-10-06.