Hacker News Reader

Comment by IncRnd | Hacker News Reader

The SHAttered attack was made public shortly after browsers deprecated SHA-1 or started flagging SHA-1 in certs as insecure. This specific practical attack may have existed for a while before it was made public. In fact it may have been made public only when it no longer provided practical benefits.

Just some food for thought.

View on HN

The team that found it is at least partly academic; they published a paper on a free-start collision at CRYPTO 2015, and made a prediction there about how long it would take them and how much it would cost to find a full collision, which was more or less accurate. They are hash function researchers, not spies or hackers. There's no reason to believe they were hiding anything.

That doesn't make a lot of sense to me, if anything this type of attack only gets more practical as the cost to compute a collision dwindles. It's not like a 0-day vulnerability.

I'm also not aware of any significant recent attempt at mitigating SHA-1 collisions that would've suddenly made the attack impractical. And in general these collisions are pretty difficult to exploit since you can't control the collision exactly.

Weaknesses with theoretical attacks have been known for over 10 years, and most major companies started phasing it out of use years ago. The recent news that someone was able to find two colliding inputs is just yet another red flag against continuous use, but there is no more need to panic now than a year ago. Remember that finding a collision for a spesific input is much harder than finding two inputs which collide.

Maybe. But SHA-1 is still widely used today, perhaps no long for signing certificates, but a lot of developers and systems today continue to use SHA1 to produce checksum.

We don't know what tools the NSA have in their arsenal.

Reply on news.ycombinator.com