Project Glasswing: Claude Mythos Preview Announced As Anthropic’s Cybersecurity AI

Anthropic has announced a new initiative called Project Glasswing, aimed at strengthening cybersecurity using artificial intelligence. The project is powered by a new general-purpose model called Claude Mythos Preview, which the company describes as its most advanced yet for coding and agentic tasks.

For now, the model is not publicly available. Instead, it is being rolled out to a select group of partners, where it will be used to identify vulnerabilities and security flaws across software systems.

ALSO READ: UIDAI Aadhaar Services: How To Generate Virtual ID For Your Card, Step By Step Guide

Backed By Major Tech And Financial Players

The initiative has launched with a wide set of partners, including Amazon Web Services, Apple, Broadcom, Cisco, CrowdStrike, Google, JPMorgan Chase, Linux Foundation, Microsoft, Nvidia, and Palo Alto Networks.

According to the company, Project Glasswing is positioned as an effort to secure critical software infrastructure, especially systems that are widely used across industries.

Claude Mythos Preview At The Core

At the centre of the initiative is Claude Mythos Preview, an unreleased AI model designed to work with complex codebases. Anthropic says the model can analyse software, detect vulnerabilities, and even help fix them. The company claims it has already identified thousands of high-severity issues, including ones affecting major operating systems and web browsers.

Beyond the initial partners, access is also being extended to over 40 organisations that manage or maintain critical infrastructure. These groups are expected to use the model as part of their defensive security efforts.

Funding And Access

To support the initiative, Anthropic has committed up to $100 million in usage credits, along with $4 million in direct funding for open-source security organisations.

Even though the model is currently limited to partners, it is not free. Claude Mythos Preview is priced at $25 per million input tokens and $125 per million output tokens. It can be accessed through multiple platforms, including Anthropic’s API, Amazon Bedrock, Google Cloud Vertex AI, and Microsoft Azure AI Foundry.

ALSO READ: Shark And Ninja Enter India With Smart Appliances: Blender, Air Fryer, HydroVac And More| Details Inside

Not Public Yet

Anthropic has made it clear that there are no immediate plans to release the model widely. The company says it wants to ensure strong safeguards are in place before scaling access.

“We do not plan to make Mythos Preview generally available. Our goal is to deploy Mythos-class models safely at scale, but first we need safeguards that reliably block their most dangerous outputs,” the company said.