I’ve used a liberated X200 since February 2018, and coming up on the first anniversary has started me thinking - was my choice to switch to the x200 a rational one?

I’ve been very worried about potential exploits involving the Intel ME (IME) and that was the motivation to use this machine. Let’s argue the other side of this: that this choice was irrational.

First, physical access to my workstation is total access. Previously:

This machine is still vulnerable to attacks that require physical access (Evil Maid, a hardware keylogger, cold boot RAM read). I would notice if the original Lenovo BIOS was flashed back on, but the X200 remains vulnerable if someone has physical access and flashed a compromised coreboot. I usually rule out attacks requiring that level of physical access (an hour or more with my machine without my knowledge) when making security related decisions. This may sound like a cop-out but I’ll reiterate: security is about tradeoffs.

Nothing I’ve done so far prevents this vulnerablility. Physical access is total access. I did verify that the sha256 of my coreboot ROM has not changed since I first flashed it during the most recent rebuild and reflash, but I don’t do this every boot. If physical access is on the table, nothing is secure.

What about a root exploit? If they have root, they have my machine. It would be less work to just compromise my browser, exfil files, install a keylogger, or something else than it would be to flash an IME exploit. I have never used the TPM for storing the keys for disk encryption so there isn’t a concern there. In systems where there is an internal programmer for the SPI chip, root is required anyway.

Ok, so say I go out and buy a new machine? The first thing I would do is run me_cleaner no matter what machine it was. I would feel a lot better with coreboot + a neutralized ME, but is coreboot required?

Say I was concerned about a UEFI exploit like LoJax. I’m pretty sure modifying UEFI modules also requires root, so we’re back to the previous argument!

I’d feel a lot better using a machine that was supported by coreboot but I’m not even sure this is necessary anymore.

So:

• a root level exploit is bad, and more likely
  • compromised browser is more likely
  • UEFI exploit is rare but has happened
    • probably needs root to modify UEFI stuff
• a IME exploit is super bad, but less likely
  • mitigated by me_cleaner soft or hard disable (?)

At this point I’m considering my previous position irrational, and have switched back to magnus for now. I ran me_cleaner on it and it was successful (although I had to soft-disable only due to Bootguard):

I’ve been ignoring OS / software level security hardening by focusing on the IME. To be clear, the X200 is more secure in that it does not have an IME and therefore cannot even be considered in a doomsday scenario, but, doomsday notwithstanding, any machine that has had me_cleaner run on it can be considered secure.

Discuss on Hacker News