RSA Attacks - Talk at Stanford - 1/28/2002 4PM (fwd)

2002-01-25 Thread Bill Stewart

Looks like an interesting talk!

-- Forwarded message --
Date: Thu, 24 Jan 2002 16:52:35 -0800 (PST)
From: Glenn Durfee [EMAIL PROTECTED]
Subject: Ph.D. Oral Exam: Monday, January 28, 4PM

Algebraic Cryptanalysis
 Glenn Durfee

Department of Computer Science
  Stanford University
Gates Building, Room 498
Monday, Jan. 28th, 2002
   4:00 PM - 5:00 PM


In this talk we study the security of the widely-used RSA public key
cryptosystem.  RSA is used in the SSL protocol for security on the
Internet, and the SET protocol used by Visa for secure credit card
transactions.  This talk outlines several cryptanalytic results on the RSA
public key cryptosystem and variants.  We obtain our results using tools
from the theory of integer lattices.

We begin by introducing a novel algorithm for the factorization of a
class of integers related closely to RSA moduli, showing a new class
of integers can be efficiently factored.  We go on to introduce
new attacks on the RSA public key cryptosystem which take advantage of
partial knowledge of a user's secret key, showing that in low public
exponent RSA, leaking the quarter least significant bits of the secret key
is sufficient to compromise RSA.  Similar results (though not as strong)
hold for larger values of the public key.  Next we describe a new attack on
the RSA public key cryptosystem when a short secret exponent is used,
extending previous bounds for short secret exponent vulnerability.  Lastly,
we describe the Sun-Yang-Laih RSA key generation schemes, and introduce
attacks to break two out of three of these schemes.

++
| This message was sent via the Stanford Computer Science Department |
| colloquium mailing list.  To be added to this list send an arbitrary   |
| message to [EMAIL PROTECTED]  To be removed from this list,|
| send a message to [EMAIL PROTECTED] For more information,|
| send an arbitrary message to [EMAIL PROTECTED] For directions|
| to Stanford, check out http://www-forum.stanford.edu   |
+-xcl+




-
The Cryptography Mailing List
Unsubscribe by sending unsubscribe cryptography to [EMAIL PROTECTED]



aibo and the dmca

2002-01-25 Thread Jim Cheesman

Apologies if this has already been sent - I don't think so, but...


Jim




From:

http://www.scientificamerican.com/explorations/2002/012102aibo/






Aibo, the Sony Corporation's popular robot dog, has delighted scores of
critics and consumers since its introduction. But the plastic pup has also
caused its creators some grief. Sony is currently struggling to resolve a
copyright dispute that centers on the work of a quirky hacker known only as
AiboPet. The controversy poses serious questions about the proper use of
robots in homes and exposes a potentially stifling effect of the U.S. Digital
Millennium Copyright Act (DMCA) of 1998.

The copyright at the heart of the case protects Aibo's encrypted brain.
AiboPet violated that copyright when he cracked the robot's source code to
reverse-engineer software that allows Aibo owners to teach their pets to
dance, speak, obey wireless commands and share the color video that serves as
their vision, among other things. None of the programs are usable without
Sony hardware and software. They earned AiboPet no money. He never revealed
the encryption code or the program he used to defeat it. Still, because the
DMCA makes it illegal to break any encrypted digital code, AiboPet's actions
made him a criminal. The fun began when Sony decided to treat him like one.



Image: Kate Wong

AIBO, Sony's robotic dog, can acquire new talents through software developed
by Sony or by a hacker known only as AiboPet.

On October 26 this past year Sony Entertainment Robots America sent AiboPet a
letter notifying him that he had violated the DMCA and demanding that he
remove all software based on their proprietary code from his Web site.
AiboPet complied immediately-and complained loudly. As a result, thousands of
Sony's best customers organized a boycott of Sony products and besieged the
company with petitions.

It is not uncommon for Aibo enthusiasts to own several of the
sophisticated-and expensive-robot dogs and to snap up each addition to a
steadily increasing library of official Aibo software at about $150 each. One
Aibo owner added his name to an open letter of protest with this addendum:
If it had not been for AiboPet's information, his invaluable knowledge and
his generosity in sharing it with the Aibo community, I would not have
purchased an Aibo, all the various software, [memory] sticks and yes, even my
computer, a Sony VAIO, which I only purchased because of its stick reader.
Another wrote: This is outrageous! I cancelled my memory sticks that were on
order. Tonight I was also going to order the wireless LAN and software...but
I have now decided not to. Also I was going to order the 310 [model] for
Christmas for my son, but I will no longer do that! I will be spending my
money elsewhere, but never again with Sony!

One of AiboPet's biggest hits was the program that teaches Aibos to two-step.
Like Kevin Bacon in the movie Footloose, he won gratitude for the gift of
dance-and Sony discovered what the town fathers in that film learned:
forbidding dancing is really, really bad for public relations.

The DMCA and Robodogs



Image: Courtesy of AiboPet

ROBODOG PACKS, including this one owned by AiboPet, are common among Aibo
enthusiasts.

The intensity with which Aibo owners reacted produced what AiboPet calls a
pretty major shift in Sony's position. By November 23rd, Sony and AiboPet
had reached an understanding that allowed the hacker to repost on his Web
site most of the programs that Sony initially asked him to remove. The
reaction had a major influence, AiboPet says, or at least I believe it
did. Sony spokespeople have refused to comment on the dispute beyond a
carefully worded statement thanking Aibo enthusiasts and pointing out that
they were only trying to protect their copyright. Sony and AiboPet continue
to work out the ground rules of their relationship, but things are going
smoothly. The boycott by Aibo lovers ended.

Still, the world may not be a safe place for teaching robot dogs new tricks.
The DMCA remains the law of the land, and what AiboPet does breaks it. Sony
retains its right to crack down on AiboPet and others like him, but chooses
not to exercise it, for now. Were AiboPet reverse-engineering any product
that did not include digital encryption to produce compatible products, he
would probably not be breaking the law. In the past, courts have upheld the
right of programmers to create games compatible with game consoles like
Nintendo and Sega. Compatible product development speeds technological
innovation, but the DMCA criminalizes that pursuit wherever encryption is
used. The Electronic Frontier Foundation, the premier Internet First
Amendment watchdog, has expressed its concern, as has the Robotics Society of
America.

The intensely personal nature of owners' relationships with their Aibos
injects real passion into this controversy and points the way for things to
come. The Sony Corporation deliberately shaped their foray into home robotics
to foster 

Re: biometrics

2002-01-25 Thread Jaap-Henk Hoepman


As much as i have my doubts about biometric systems i cannot let the below
pass. 

On Wed, 23 Jan 2002 21:11:23 +0100 Perry E. Metzger [EMAIL PROTECTED] writes:
 However, as soon as you lose physical control over the device doing
 the measurements or their communications path biometrics become worse
 than useless. As one example, they're useless for authenticating
 over-the-net bank account access -- the device on your desk that your
 bank helpfully provides to scan your eye might not even be attached
 when the cracker's software helpfully provides forged information down
 the line. Liveness tests are not useful if you don't even know if
 the biometric hardware at the other end is intact. Anything in a
 user's location is by definition untrustworthy in this sense.

Of course (and i think Dorothy mentioned this too), the measuring device and
it's connection to the veryfying system must be properly protected. In case of
the system Perry describes, a secure and fresh (ie fresh session key) link
should be setup between the measuring device and the bank, so that
eavesdropping _and_ replay/forgery is impossible. Even though most biometric
systems may not implement this (i simply don't know), this is not a weakness of
biometric systems per se.

[Moderator's note: er, HUH? How does the link being realtime assure
that the remote side isn't simply generating iris images and sending
them to you? It doesn't. Biometrics are worthless except when the
entire system is completely physically secure. --Perry]

Jaap-Henk
 
-- 
Jaap-Henk Hoepman | Come sail your ships around me
Dept. of Computer Science | And burn your bridges down
University of Twente  |   Nick Cave - Ship Song
Email: [EMAIL PROTECTED] === WWW: www.cs.utwente.nl/~hoepman
Phone: +31 53 4893795 === Secr: +31 53 4893770 === Fax: +31 53 4894590
PGP ID: 0xF52E26DD  Fingerprint: 1AED DDEB C7F1 DBB3  0556 4732 4217 ABEF




-
The Cryptography Mailing List
Unsubscribe by sending unsubscribe cryptography to [EMAIL PROTECTED]