4 ms·

Why is TikTok hiring quantum physics researchers?

2 points by skanderbm 4y ago

pyb 4y ago
Either an internal joke, or a token research program to satisfy the Chinese government ! https://zox-news.com/less-tiktok-more-quantum-in-china-a-tech-sector-under-control/ https://zox-news.com/less-tiktok-more-quantum-in-china-a-tec...
abudabi123 4y ago
We live in the Quantum Information-age of cyberphysical. Data for dummies is not for TikTok?
- dontbenebby 4y ago
  What are the implications of the "Quantum Information-age of cyberphysical"? I had a conversation about "quantum" with a coworker who was an ex-NSA intern around the same time as the 2009 G20, the main takeaways I got were to keep things offline as much as possible, and that symmetric ciphers like AES would remain as strong as their keys but I don't fully grok the practical implications of how a company like Tik Tok (which is an extension of the PRC by virtue of being a Chinese company) would use such knowledge.
  - raxxorraxor 4y ago
    I think the scare of quantum computing was that it could solve an nP problem in polynomial time. Encryption uses "lock-in functions" that can be verified quickly, but not solved quickly. Prime factorization for RSA or finding the inverse function for EDDSA. Don't know where the idea first surfaced, but Grovers algorithm is an example where quantum computing can have an advantage compared to conventional machines: https://en.wikipedia.org/wiki/Grover%27s_algorithm https://en.wikipedia.org/wiki/Grover%27s_algorithm If true, our lock-in functions for encrypted data might need to be revised because a quantum computer could solve them faster.
    - dontbenebby 4y ago
      Thanks... I lack the math background to grok "ould solve an nP problem in polynomial time." without doing some refreshing, but I really appreciate you distilling that into one plain sentence. (I never took Calc, I'd get impatient because I skipped geometry and trig and a few other things before I went off to uni.) >If true, our lock-in functions for encrypted data might need to be revised because a quantum computer could solve them faster. Yikes, though it was my understanding symmetric ciphers (ex: AES) wouldn't be as impacted. I tried to find a cite, and all I found was an attack paper backing up the assertion previous attacks were mostly on asyymetric ciphers. (Not a cryptographer, so I often can just repeat the takeaways from what others explained in detail their reasoning on): > So far, in cryptography, the impact of quantum computing is analyzed mostly for asymmetric cryptography and less is known about its consequences on symmetric cryptography [0] https://www.sciencedirect.com/science/article/abs/pii/S0045790622003743 https://www.sciencedirect.com/science/article/abs/pii/S00457... I try to keep things offline, but it's hard to keep up with all these... heuristics :-)
dbrix 4y ago
The JD makes non-sense. This link makes more sense https://jobs.bytedance.com/en/position/7054987796163774734/detail?spread=BSPP2KS https://jobs.bytedance.com/en/position/7054987796163774734/d....

2 points by skanderbm 4y ago

pyb 4y ago

Either an internal joke, or a token research program to satisfy the Chinese government ! https://zox-news.com/less-tiktok-more-quantum-in-china-a-tech-sector-under-control/ https://zox-news.com/less-tiktok-more-quantum-in-china-a-tec...

abudabi123 4y ago

We live in the Quantum Information-age of cyberphysical. Data for dummies is not for TikTok?

dontbenebby 4y ago
What are the implications of the "Quantum Information-age of cyberphysical"? I had a conversation about "quantum" with a coworker who was an ex-NSA intern around the same time as the 2009 G20, the main takeaways I got were to keep things offline as much as possible, and that symmetric ciphers like AES would remain as strong as their keys but I don't fully grok the practical implications of how a company like Tik Tok (which is an extension of the PRC by virtue of being a Chinese company) would use such knowledge.
- raxxorraxor 4y ago
  I think the scare of quantum computing was that it could solve an nP problem in polynomial time. Encryption uses "lock-in functions" that can be verified quickly, but not solved quickly. Prime factorization for RSA or finding the inverse function for EDDSA. Don't know where the idea first surfaced, but Grovers algorithm is an example where quantum computing can have an advantage compared to conventional machines: https://en.wikipedia.org/wiki/Grover%27s_algorithm https://en.wikipedia.org/wiki/Grover%27s_algorithm If true, our lock-in functions for encrypted data might need to be revised because a quantum computer could solve them faster.
  - dontbenebby 4y ago
    Thanks... I lack the math background to grok "ould solve an nP problem in polynomial time." without doing some refreshing, but I really appreciate you distilling that into one plain sentence. (I never took Calc, I'd get impatient because I skipped geometry and trig and a few other things before I went off to uni.) >If true, our lock-in functions for encrypted data might need to be revised because a quantum computer could solve them faster. Yikes, though it was my understanding symmetric ciphers (ex: AES) wouldn't be as impacted. I tried to find a cite, and all I found was an attack paper backing up the assertion previous attacks were mostly on asyymetric ciphers. (Not a cryptographer, so I often can just repeat the takeaways from what others explained in detail their reasoning on): > So far, in cryptography, the impact of quantum computing is analyzed mostly for asymmetric cryptography and less is known about its consequences on symmetric cryptography [0] https://www.sciencedirect.com/science/article/abs/pii/S0045790622003743 https://www.sciencedirect.com/science/article/abs/pii/S00457... I try to keep things offline, but it's hard to keep up with all these... heuristics :-)

dbrix 4y ago

The JD makes non-sense. This link makes more sense https://jobs.bytedance.com/en/position/7054987796163774734/detail?spread=BSPP2KS https://jobs.bytedance.com/en/position/7054987796163774734/d....