4 ms·

It's a thought experiment on the viability of retrocausal blackmail, a techno-futurist rebranding of conventional christian eschatology. The real-life hazard is

It's a thought experiment on the viability of retrocausal blackmail, a techno-futurist rebranding of conventional christian eschatology. The real-life hazard is minimal, although probably not strictly zero. There are known real informational hazards for humans, but they generally only affect people with epilepsy or those predisposed to staring at odd optical illusions for longer periods of time than most would consider reasonable.

silicon2401 5y ago
Do you have any examples or references for real informational hazards for humans? I'd love to read more about that
- katmannthree 5y ago
  Absolutely! I'm afraid they're actually pretty boring though. The most common informational hazards are varying forms of behavioral modifiers, e.g. advertising, religions/cults, state-sponsored propaganda. Edward Bernays' is one of the more interesting people to study here, both through his written works (namely Crystalizing Public Opinion and Propaganda, both from the 1920's) and his later actions (popularizing female smoking, helping fruit companies and the CIA overthrow South American governments, etc). Roko's Basilisk falls into this category, if you want to think about it really hard it can hurt you but(somewhat ironically) the only people really susceptible are pseudo-intellectuals who congregate on places like lesswrong. Next up are simple forms of adversarial inputs such as discordant sounds or rapidly blinking lights. People's susceptibility to these varies, generally only those with neurological disorders (epilepsy, severe misophonia) are severely impacted by known tech. Applied examples include LRADs, dazzlers[0], and/or some bowling alleys. Last up there is at least one true informational hazard where perception alone is enough to disrupt the underlying structure of your brain (or at least your visual cortex). It's called the McCollough effect, and I'm not going to link to it but if you're curious a quick glance is generally safe. On the surface it's a simple color shifting optical illusion, but unlike most of those, if you stare at it for more than a handful of seconds, the aftereffect lasts for months. [0]: https://en.wikipedia.org/wiki/Dazzler_(weapon) https://en.wikipedia.org/wiki/Dazzler_(weapon), the boring ones are based on just blinding while the interesting ones attempt to disorient your brain with high frequency color patterns
  - silicon2401 5y ago
    Thanks for the reminder, I've read about Bernays before (maybe in the Century of the Self documentary?) and forgot that I've been meaning to study him more. I think behavioral modifiers is a great example. That's the exact kind of thing I was hoping for; something that reveals vulnerabilities we have but feel like we shouldn't. I also love the adversarial inputs and McCollough effect. Another example of what I mean: it shows how we're fundamentally just machines that have exploits like any other, and in a very dangerous way. Thanks for sharing; I find stories and topics that try to elevate and glorify mankind tired and repetitive. In contrast I find things that reveal the dark sides and weaknesses of humankind endlessly fascinating. Unfortunately those topics don't seem to be as popular to create content about.
    - katmannthree 5y ago
      I find these types of things to be a lot more interesting as well. With respect the behavioral modifiers I feel like they really help explain how societies actually function, once you start paying attention you see just how effective and pervasive they really are. This study[0] discusses adversarial images which fool both ML classifiers and time limited humans, a rather interesting development in the field. It's a good space to keep an eye on in general, at some point someone's probably going to discover a snowcrash-level informational hazard and it would be nice to have some warning. [0]: https://arxiv.org/pdf/1802.08195.pdf https://arxiv.org/pdf/1802.08195.pdf

silicon2401 5y ago

Do you have any examples or references for real informational hazards for humans? I'd love to read more about that

katmannthree 5y ago
Absolutely! I'm afraid they're actually pretty boring though. The most common informational hazards are varying forms of behavioral modifiers, e.g. advertising, religions/cults, state-sponsored propaganda. Edward Bernays' is one of the more interesting people to study here, both through his written works (namely Crystalizing Public Opinion and Propaganda, both from the 1920's) and his later actions (popularizing female smoking, helping fruit companies and the CIA overthrow South American governments, etc). Roko's Basilisk falls into this category, if you want to think about it really hard it can hurt you but(somewhat ironically) the only people really susceptible are pseudo-intellectuals who congregate on places like lesswrong. Next up are simple forms of adversarial inputs such as discordant sounds or rapidly blinking lights. People's susceptibility to these varies, generally only those with neurological disorders (epilepsy, severe misophonia) are severely impacted by known tech. Applied examples include LRADs, dazzlers[0], and/or some bowling alleys. Last up there is at least one true informational hazard where perception alone is enough to disrupt the underlying structure of your brain (or at least your visual cortex). It's called the McCollough effect, and I'm not going to link to it but if you're curious a quick glance is generally safe. On the surface it's a simple color shifting optical illusion, but unlike most of those, if you stare at it for more than a handful of seconds, the aftereffect lasts for months. [0]: https://en.wikipedia.org/wiki/Dazzler_(weapon) https://en.wikipedia.org/wiki/Dazzler_(weapon), the boring ones are based on just blinding while the interesting ones attempt to disorient your brain with high frequency color patterns
- silicon2401 5y ago
  Thanks for the reminder, I've read about Bernays before (maybe in the Century of the Self documentary?) and forgot that I've been meaning to study him more. I think behavioral modifiers is a great example. That's the exact kind of thing I was hoping for; something that reveals vulnerabilities we have but feel like we shouldn't. I also love the adversarial inputs and McCollough effect. Another example of what I mean: it shows how we're fundamentally just machines that have exploits like any other, and in a very dangerous way. Thanks for sharing; I find stories and topics that try to elevate and glorify mankind tired and repetitive. In contrast I find things that reveal the dark sides and weaknesses of humankind endlessly fascinating. Unfortunately those topics don't seem to be as popular to create content about.
  - katmannthree 5y ago
    I find these types of things to be a lot more interesting as well. With respect the behavioral modifiers I feel like they really help explain how societies actually function, once you start paying attention you see just how effective and pervasive they really are. This study[0] discusses adversarial images which fool both ML classifiers and time limited humans, a rather interesting development in the field. It's a good space to keep an eye on in general, at some point someone's probably going to discover a snowcrash-level informational hazard and it would be nice to have some warning. [0]: https://arxiv.org/pdf/1802.08195.pdf https://arxiv.org/pdf/1802.08195.pdf