[7557] in cryptography@c2.net mail archive
Re: A proposal for secure videoconferencing and video messaging
daemon@ATHENA.MIT.EDU (eli+@gs211.sp.cs.cmu.edu)
Wed Jul 26 14:45:55 2000
From: eli+@gs211.sp.cs.cmu.edu
Message-Id: <200007261420.HAA13648@blacklodge.c2.net>
To: crypto list <cryptography@c2.net>
Date: Wed, 26 Jul 2000 10:05:20 -0400 (EDT)
In-Reply-To: <4.3.1.2.20000725085627.02175d78@shell11.ba.best.com> from "James A. Donald" at Jul 25, 2000 08:59:52 AM
Reply-To: eli+@cs.cmu.edu
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
James A. Donald wrote:
> With video, the authentication problem, which has always been the great
> barrier to the widespread use of crypto, goes away.
>
> [Does it? In a few years, it should be possible to synthesize video in
> real time... --Perry]
Today's best synthesized voice is clearly not human at all, much less
any particular human, but VOIP hasn't made authentication go away yet.
Real-time synthesis of face and body video may be easier or harder
than that of voice, I don't know, but for some time I think the best
attack against each will be old-fashioned analog impersonation.
That's a fair bit harder for video/audio than for audio alone, maybe
hard enough for some authentication needs.
--
Eli Brandt | eli+@cs.cmu.edu | http://www.cs.cmu.edu/~eli/
