PCI-grade hardware wallets.

[satscraper]

The wallets listed below are known to use tamper‑resistance concepts similar to those adopted by  Payment Card Industry, i.e. they implement various techniques that help to resist probing of their SE:

• Keystone Pro(complex circuitry that secures self‑wipe on tamper, RAM reset at sudden environment change, scrambling "sensitive information leaked out by power consumption or electromagnetic emissions", etc..)
• NGRAVE ZERO("sensors inside NGRAVE ZERO which wipe the device if opened".)
• Ellipal Titan*("automatically deletes the device's data in case of a breach making the device unusable".)

*Partially.

Further updates will follow.

Please feel free to help refine or expand the list.

[dkbit98]

I wouldn't trust any closed source devices like Ellipal Titan with their claims, especially knowing their history of exploits.
Another case in history I remember is when one guy opened Keystone device and there was not self-wipe happening.
I didn't investigate this subject deeper but I am sure Satochip and Keycard are using something similar, since they are based on secure elements in java cards, just like in payment cards.
This ''protection'' doesn't really exist unless it is well documented and confirmed by independent testing.

 

[Pmalek]

This ''protection'' doesn't really exist unless it is well documented and confirmed by independent testing.

That's just the thing. I don't think there are many people who research these tamper‑resistant features in hardware wallets and document their findings. Some companies post about them and make claims, but where is the proof behind those pretty words? The industry would need teams like Donjon to carry out the research or independent researchers or hackers to take a look and write about their findings.