Fujitsu (claims to have) established a new record "smashing" pairing-based cryptography at 923 bits with 252 cores in 148.2 days, while the previous record in 2

Fujitsu (claims to have) established a new record "smashing" pairing-based cryptography at 923 bits with 252 cores in 148.2 days, while the previous record in 2009 was 676 bits. The Fujitsu announcement is here: http://www.fujitsu.com/global/news/pr/archives/month/2012/20120618-01.html http://www.fujitsu.com/global/news/pr/archives/month/2012/20... Can anyone find a research paper on that? Anyway, IMHO the article posted looks a bit confuse (and confusing...) For pairing-based cryptography (although precise settings are unclear), 923-bit security is still kind of poor: less than 1024, which is considered equivalent to 80-bit symmetric security. This means that AES-128 is still far away... I don't really get the relationship with AES.