Researchers discover security flaw in microchips

[upi402]

"Security has been sacrificed for the benefit of performance," French daily Le Monde quoted Seifert as saying Saturday.

The danger of hackers breaking into computer chips, using a technique known as Branch Prediction Analysis (PBA), previously necessitated a very large number of attempts to decipher a cryptology key.

But Seifert's research envisages that hackers will find it far easier and quicker in the future using a small piece of spy software that tracks microchips.

In a still-confidential study, Seifert and his colleagues explain how they managed "in only one attempt" to obtain a 512-bit encryption key in just a few thousandths of a second.

http://www.physorg.com/news83072716.html

[cosmik debris]

1. Does this mean the end of PGP? n/t

[Irreverend IX]

2. Hackers breaking into computer chips? haha

Another technology writer who obviously knows nothing about technology. I'm interested in knowing how you'd "break into" a solid-state hunk of silicon. I think the study's actual finding is that hackers have found a way to predict what sort of cryptography keys certain processors will create. I believe that producing secure encryption keys requires the CPU to create some random numbers, and chipmakers cut corners in such a way that their chips spit out easily predictable, quasi-random numbers, which allows encryption keys to be quickly cracked by someone who has a program that predicts the numbers a given chip will produce.

[Kagemusha]

3. Sounds like you're right...

The article's just poorly presenting the point.

[upi402]

4. Thanks, I'm no expert but was curious

if this was a valid concern.

So what is the current, best security/privacy? Is PGP still it?
For me it needs to be user-friendly.

[Irreverend IX]

5. I don't think the program you use makes a difference...

It's the processor you have. I'd think that as long as your processor can produce relatively unpredictable numbers, you'll be able to create secure PGP keys. But the article says nothing about which processors have the problem and which don't, so there's no way to know yet whether or not your CPU can create safe keys.