Recent searches
Search options
my 2025 infosec prediction is that asset lists will still be as outdated and incomplete and nonexistent as we saw in 2005.
here are some more... 1/??????
and also source code access will continue to be denied to assessors despite source assisted tests being vastly greater value for all involved
ransomware groups will continue to absolutely shithouse major companies using really low-hanging techniques and legislators will blather on about doing something about it for another 12 months unless maybe someone says china did it and then racism might get them off their ass and we'll all be deeply disillusioned about it
Bloomberg will continue to have the same highly accurate and well informed technical reporting as they have displayed in prior years
on the plus side there will be some funny vulnerability names
also someone will overhype the absolute shit out of a bug that turns out to be a complete nothingburger. I'm gonna guess... something related to USB-C.
someone will try to make a thing that competes with Signal and it will go very badly and soatok will begrudgingly write a blog post about it
UPDATE: just 15 days into 2025 and it came true: http://soatok.blog/2025/01/14/dont-use-session-signal-fork/
Microsoft will get owned using creds stolen from Recall on one of their staff's systems
(this one is less of a prediction and more of a "please I need this to happen because it would be a gem of pure amusement in an otherwise cold and indifferent world")
IBM will announce a blockchain security product
Quantum computers will be able to factor a bigger number but it will still be small enough that you could validate the result on a $5 calculator. The tech press will claim it breaks AES.
@gsuberland For this one to be true, they'd have to have factored any number at all already. AIUI they've only "factored" fixed inputs baked into the program tailored to the specific input, which isn't even computation at all. It's like saying printf("2,3"); factored 6.
@dalias @gsuberland I thought they had managed to actually factor arbitrary numbers up to like 20 by this point?
@azonenberg@ioc.exchange @dalias@hachyderm.io @gsuberland@chaos.social recently came out to like ~20 bits, but that's still laughable to what you can do on a phone (~400bits at least) https://www.csoonline.com/article/3562701/chinese-researchers-break-rsa-encryption-with-a-quantum-computer.html
@ignaloidas @gsuberland @azonenberg Not clear if the claim is truthful or another case of "chosen input factoring".
@dalias@hachyderm.io @gsuberland@chaos.social @azonenberg@ioc.exchange They say that they have factored a bunch of randomly chosen ones? But also can't really say, the paper is in chinese, only the abstract being in english
@ignaloidas @gsuberland @azonenberg I mean factoring arbitrary 22 bit numbers is a 44Mbit lookup table so 
@dalias@hachyderm.io @gsuberland@chaos.social @azonenberg@ioc.exchange yeah, it still needs to get hundreds of orders of magnitude larger to get to sizes that would be worrying, which will take a long time without any sudden breakthroughs