Fifth Third Bancorp has been invited to join Anthropic’s Project Glasswing, a pivotal AI cybersecurity initiative, as reported by Seeking Alpha on Wednesday (June 10). The announcement was made by Bryan Preston, Fifth Third’s chief financial officer, at the Morgan Stanley U.S. Financials Conference, confirming the bank gained access to the program within the past several weeks.
Preston emphasized the strategic importance of Fifth Third’s role in the national financial infrastructure as a key factor in its selection. “We think it was a reflection of just the role we play in the payments ecosystem in the country today, whether it’s the Direct Express business, some of the processing that we do for U.S. Customs as well as just the magnitude of payroll processing that we do for the country,” Preston stated, according to the report. This underscores the bank’s significant involvement in critical financial operations, including its role as the financial agent for the U.S. Treasury Department’s Direct Express prepaid debit card program, a five-year agreement inked last September.
Project Glasswing’s Genesis and Impact
Anthropic initially unveiled Project Glasswing in April, coinciding with the limited release of its first Mythos-class AI model, Claude Mythos Preview. The initiative was designed to grant select partners early access to this advanced AI model, enabling them to leverage its sophisticated cybersecurity capabilities to fortify their own systems ahead of a broader public release.
The early results of Project Glasswing have been substantial. By May 22, Anthropic reported that Claude Mythos Preview had successfully identified more than 10,000 cybersecurity vulnerabilities. These critical findings were made in what the company described as “the most systemically important software in the world,” allowing for timely patching and remediation.
Expansion and Advanced AI Deployment
The success and strategic importance of Project Glasswing led Anthropic to announce its expansion on June 2. Initially, the cybersecurity effort provided approximately 50 organizations with access to Claude Mythos Preview. This has now been significantly scaled up to include 150 organizations, broadening the initiative’s reach and impact across various sectors.
Further enhancing the program, Anthropic introduced two new Mythos-class models on Tuesday (June 9), developed with robust safeguards to prevent misuse. One of these, Claude Mythos 5, is slated for initial release exclusively through Project Glasswing, notably in collaboration with the U.S. government. This strategic partnership highlights the critical nature of the AI’s application in national cybersecurity efforts.
Fifth Third’s participation in Project Glasswing comes as the bank recently solidified its position in the U.S. financial landscape. On February 2, Fifth Third became the ninth-largest U.S. bank by assets following the successful merger between Fifth Third Bancorp and Comerica. This enhanced scale and systemic importance likely contributed to its invitation to Anthropic’s cutting-edge cybersecurity program, positioning it at the forefront of AI-driven defense against evolving digital threats.
