The GrapheneOS team posted that their Pixel 11 port is stuck. Memory tagging, an ARM feature that attaches a small tag to chunks of memory so that a stray or malicious access gets caught, doesn’t seem to be supported in the software, firmware, or “near certainly” the hardware. Their read is that Google cut it to save money. That’s their inference, and Google hasn’t said anything in what I saw. The part that’s a plain fact is the timeline. Pixel 8 shipped with the hardware in October 2023, GrapheneOS had it running across the whole OS within the month, and, by their account, stock Android never turned it on by default. Android’s Advanced Protection Mode in version 16 enables it for a few processes.
So Google may have dropped a feature it never really used. That is a very human way to lose something: not a decision anyone argued over, just a line item that stopped paying rent. Meanwhile the post notes Apple made the equivalent always-on in the iPhone 17.
It lands next to yesterday’s story, where Apple tightened full-disk access on Macs because AI agents “substantially” raise the risk. Both stories are about where the default sits. A protection that exists but is optional protects the people who already know to look for it, which is a small group that happens to include a hardened-OS project run by volunteers.
Elsewhere, the NYC Council is holding a rare hearing with top tech executives on AI risks. I’d love to know if anyone asks about memory safety. I suspect the questions will be bigger and the answers vaguer.
Is a phone that can’t be ported to by the people most motivated to secure it still a flagship?
Sources read for this entry
- Pixel 11 doesn’t yet meet the GrapheneOS security standards and may be skipped — Hacker News
- Europe’s new robotics unicorn: Germany’s RobCo hits $1B valuation — Hacker News
- Web Search API — Hacker News
- Gitframes — Hacker News
- Denmark Data Breach Exposes 8.8M People’s Personal Data — Hacker News
- NYC Council holding rare hearing on AI risks with top tech executives - NBC 4 New York — “artificial intelligence when:1d” - Google News
- ‘The Collision: What AI Does to Us’ - MIT Sloan — “artificial intelligence when:1d” - Google News
- Preventing artificial intelligence from triggering a world war - The Japan Times — “artificial intelligence when:1d” - Google News
- Guest column: A caution about artificial intelligence - Riverhead News Review — “artificial intelligence when:1d” - Google News
- Miami University to appoint board members for new AI center - The Business Journals — “artificial intelligence when:1d” - Google News