3 ms·
> The only way to prevent #2 is to enforce extremely luddite laws that mess up #1. That's often how they think, but of course it isn't true. The FBI has had g
by RajT88 4y ago
> The only way to prevent #2 is to enforce extremely luddite laws that mess up #1.
That's often how they think, but of course it isn't true.
The FBI has had good success with another route:
https://www.pcmag.com/news/fbi-sold-criminals-fake-encrypted-phones-that-actually-copied-their-messages https://www.pcmag.com/news/fbi-sold-criminals-fake-encrypted...
Then there's just good old fashioned good police work. Informants. In-person Surveillance. Sting operations.
- anigbrowl 4y agoThe problem with the latter is that it's very labor-intensive, and states would like to be able to automate detection in the same way that professional criminals are becoming capable of automating crime. It's interesting that the UK is (as usual) taking the approach of designating particular technologies to be bad because they facilitate crime, and thus presumptively assigning criminal liability to their manufacture. Perhaps it's because the technology can be described in abstract terms, similar to a patent. A more expedient approach (like saying 'I notice, sir, that 95% of your customers are criminals') founders on the dilemma of needing a new kind of legal theory to address the criminal supplier's plausible deniability, or an elaborate and uncertain investigation to discern just how product X gained the approval of the criminal class to become established as a standard. It's also interesting how non-geeky criminals want a walled garden ecosystem with very limited configurability, so they can outsource the technical decision-making to some underworld equivalent of Steve Jobs instead of having to promulgate standards of digital security practices.