This is the inevitable follow up to this:

https://www.reddit.com/r/sysadmin/comments/9mk2o7/mri_disabl...

This makes me wonder if someone could somehow silently inject helium in a room to kill all phones (iPhone only?) for malicious purposes

I'd just like to point out, for folks who are thinking of trying this at home, that in addition to killing iPhones, helium has been known to kill people: https://www.news.com.au/lifestyle/health/inhaling-gas-from-h...

Helium doesn't kill people, lack of oxygen kills people. Helium also doesn't stay in the same place for long - it is too lightweight to even stay in the atmosphere. Any helium released in a room will immediately float upwards and rapidly be gone forever.

For someone to die using helium, they would literally have to first find pure helium (which isn't sold at party stores any more - helium is mixed with oxygen) and pump it into a plastic bag over their head with the bag sealed (because it will come off when you pass out, you will breath in oxygen, then wake up).

> Helium doesn't kill people, lack of oxygen kills people.

Helium embolism can kill people. Due to helium's permeability, it can travel through tissues and into the blood stream. Helium gas bubbles form in the bloodstream, preventing the flow of blood.

https://www.sciencedirect.com/science/article/pii/S134462231...

Divers routinely breathe >90% helium mixes with no ill effects provided they ascend and descend at the proper rates. I very much doubt that helium embolism is a real phenomenon at normal atmospheric pressures. More likely that article is just completely wrong based on misinterpreting postmortem changes.

If one is breathing helium directly from a pressurised tank, one can be inhaling it at higher than normal atmospheric pressure.

Divers control the pressure at which the diving gas is being delivered. A person who is breathing helium directly from a tank, either for misguided recreational reasons or with suicidal intent, is unlikely to be doing that.

The original suggestion – of adding helium to a room in order to damage iPhones – unless the room is pressurised, it is going to be at normal atmospheric pressure, so asphyxiation is a far greater risk than embolism. But, the point remains, inhaling helium is actually a lot more risky than many people think, sometimes it kills.

https://www.oregonlive.com/pacific-northwest-news/index.ssf/...

Yes, fairly easily. The trick is finding the sweet spot of concentration (speed of action vs difficulty vs detectability). As a SWAG I'd say somewhere in the 5-10% range is probably optimal.

It only disables them, doesn't kill them. After the helium has time to diffuse back out the phone starts working again.

It is conceivable that a powered chip clocked with a MEMS oscillator that is seriously off-frequency may latch into a state that is unrecoverable so long as power is applied.

Actually, this would be a practical application in a prison setting.

Previously discussed: https://news.ycombinator.com/item?id=18340693

This would make a great plot device for a murder mystery story.