Simo Ryu Shares GPT 5.6 Prime Gap Claim

Simo Ryu posted on X that GPT 5.6 produced a stronger bound on large prime gaps. The tweet states the new result improves on the 2014 Ford-Green-Konyagin-Maynard-Tao bound by a factor of log3 X over (log4 X) squared. Ryu recalls Terence Tao saying a genuine new idea would be needed to surpass that work. The post links to an erdosproblems.com forum thread titled proof claims. No verification or independent sources appear in the packet.