How BT is Using Anthropic's Frontier AI to Halt Cyberattacks

The telecommunications industry has spent years building faster networks. Now it faces a different challenge, which is defending those networks against sophisticated cyber threats.

That reality is behind the decision of BT to become the first UK company to join Project Glasswing. 

The initiative, created by AI safety and research company Anthropic, gives critical infrastructure operators access to advanced AI tools. These systems are designed to identify vulnerabilities before attackers can exploit them.

The partnership gives BT access to Claude Mythos Preview, which is the frontier AI model of Anthropic. The telecom operator intends to use the technology to strengthen cyber defences across its networks and customer services.

Allison Kirkby, CEO at BT, announced the collaboration during the UK Government’s first AI Adoption Summit. Political and technology leaders gathered at the event to discuss how AI can support economic growth and public services.

In a speech opening the summit, Allison highlighted the relationship between AI and telecoms infrastructure, arguing that digital services depend on secure connectivity.

She says: “AI only works at scale when it is underpinned by future-ready networks that are secure, resilient and safe.” 

Allison reiterated the commitment of BT to working with the UK Government to support the development of sovereign British AI capability. 

She explains that this will help the UK to be an AI maker and not just a taker. She also emphasised the role of BT as an enabler of responsible AI adoption.

Deploying machine speed security systems

Project Glasswing brings together operators responsible for essential services. The initiative enables trusted organisations to use AI systems to uncover security weaknesses and accelerate remediation efforts.

The programme aligns with the role of BT as a provider of communications infrastructure and managed security services across the UK. 

The company blocks around four million cyberattacks across its networks every day, which illustrates the volume of malicious activity targeting digital infrastructure.

As cybercriminals adopt automated tools, telecom providers face pressure to modernise security operations. Operators must reduce the time it takes to identify and respond to threats.

BT has introduced AI-powered cybersecurity services for organisations of different sizes, including products aimed at small businesses. 

The company also announced a collaboration with Accenture focused on developing advanced AI-driven cyber operations capable of responding to threats at machine speed.

According to Jon James, CEO of BT Business, joining the Anthropic initiative will help strengthen those capabilities.

He says: “AI is changing cyber security fast, and businesses need trusted partners who can help them stay one step ahead. By joining Project Glasswing, BT will strengthen its own cyber security capability to protect our networks, our customers and the wider UK.”

As operators position themselves as managed security providers, cyber resilience is becoming as important as network performance. 

This is particularly true today, as the UK Government places greater emphasis on protecting critical national infrastructure.

For telcos, one of the biggest challenges is now ensuring that the networks supporting AI economies remain secure, resilient and trusted.