> In this case, it might be better to assume malice rather than incompetence.

I really want to consider you paranoid, but sadly I strongly agree. This is hardly the first time engineered paranoia has gripped the country, but living through it is horrible.

I was a kid during the mid-to-late Cold War (post "duck and cover") and somehow I was never able to take it seriously. Even when I took classes on strategic deterrence and the like in college I considered them light entertainment I was paying for to give me a break from the serious classes.

Now we have the emperor's new suicide vest.

Unfortunately, the threat was (and still very much is) real:

Stanislav Petrov: The man who may have saved the world http://www.bbc.com/news/world-europe-24280831

Thank you Vasili Arkhipov, the man who stopped nuclear war https://www.theguardian.com/commentisfree/2012/oct/27/vasili...

Not that paranoia is a particularly useful response, but it did not require much engineering.

Engineered paranoia is still very real and a danger to our societies because of its reality-distorting effects that can result in violent overreactions.

Did I get that right?

Great point. One idea I had recently is to start an organization in the same spirit as Open AI called OpenSurveillance that builds and releases all sorts of useful tools for thwarting surveillance efforts or acting as an adversary to an entity with pervasive surveillance power (generally speaking, corrupt regimes around the world).

I think it would entail a combination of social/operational and technological tools, and would be useful for defensive and offensive efforts.

When OpenAI releases a state of the art facial recognition algorithm, it becomes obvious how the technology might be used for good or for evil. The same goal would apply.

Perhaps one utility conducts an analysis of public social media data to determine the best strategy for bribing or compromising an adversary. Another could explain how to use a HackRF YARDStick One to track the movement of an adversary via tire air pressure sensors. Another might utilize an infra-red camera to determine which entrances and exits of a building are slow to close and could offer time to sneak in.

Still other tools might be a user-friendly rootkit installer for easy installation on a spouse or boss's device, etc., or maybe a program that trains a fleet of drones to follow a person all day, causes vehicle engine failure, etc.

The basic idea would be to highlight enough about reality so that the security (and privacy) implications of the policies can be weighed accurately by the public. By putting all the tools in one place and releasing polished, thoughtful products, the organization would help the public understand the privacy/security tradeoffs much better.

Basically a modern Anarchist's Cookbook for the surveillance age.

Disclaimer: It is not the intent of this post to discourage people to break laws, simply to use their vote and influence to peacefully change laws.

I've been considering a comparable project, but more focussed on opsec/infosec in a file sharing context. However, I've felt chilled by the possible legal consequences to me and my family if I enter this arena.

Bootstrapping myself to have great opsec seems like a really tough task. I feel like I'd have to repeatedly burn hardware, houses and identities if I want to stand a chance to reach "opsec heaven" where I can freely persue projects such as the Spook's Cookbook or the Pirate's Cookbook.

I think it could but it would need good legal support and PR support to prevent its actions from being mischaracterized.

> Bootstrapping myself to have great opsec seems like a really tough task

I think it would be incredibly difficult if not impossible in today's world. So I think that an "in the light" approach is much less vulnerable.

>a user-friendly rootkit installer for easy installation >a program that trains a fleet of drones to follow a person all day, causes vehicle engine failure, etc.

I'm afraid corrupt regimes would be the ones who benefit the most from such tools.

They already do. But since the tools are under the radar there is not all that much pressure on technology firms to follow best practices.

For example, why don't we have fully auditable filesystems?

While I think (and hope) this is correct, I'm not sure it matters.

For example, it would not be necessary for most of the engineers to be aware of a backdoor or other known vulnerability. There have been examples from open source crypto where malicious code has weakened it significantly and still nobody noticed.

There's also the very real possibility of baking the backdoor/vulnerability into a custom ASIC design. Chances are the government has a lot of expertise in this area and could simply tell Apple that it would provide one of the parts for all iPhones and the part would behave to spec (but would contain other undocumented behavior).

I think it's also realistic that other governments do this. Unless a chip is manufactured using the latest microprocessor-level miniaturization, it could contain all sorts of undocumented circuitry. I'm not sure about the economics of this sort of attack, but surely it makes sense once in a while.

There's this weird denial that takes place. I see it in all professions.

Eh, communism was way more of a threat. People in extremely high positions were communist traitors (eg Harry Dexter White who negotiated Bretton Woods/the creation of IMF for the US, or Alger Hiss who was involved in the creation of the UN, or a gazillion people in the British intelligence services).

If the government today was filled with people loyal to ISIS or AQ, then you could draw such a parallel.