Anthropic expands access to Mythos AI model, including India

Anthropic announced on Tuesday that it is granting access to its newly‑launched Mythos AI model to roughly 150 organisations worldwide. The model, designed to scan code and systems for security flaws at unprecedented speed, has already uncovered thousands of vulnerabilities in early testing. Participants span over 15 countries and include operators of critical infrastructure, signaling a coordinated effort to shore up digital defenses before threats can be exploited. India features prominently in the latest expansion, as the company aims to bring the technology to local enterprises and government agencies that manage essential services.

What happened

Anthropic’s rollout of Mythos follows a limited‑beta phase in which the AI identified a large number of security gaps in software and network configurations. The company now invites a curated list of organisations to integrate the model into their vulnerability‑management pipelines. Access is being granted to entities in more than 15 nations, with a specific focus on sectors such as energy, transportation, and finance that underpin national economies. The announcement, reported by the Economic Times, confirms that Indian organisations are part of this cohort, marking the first formal entry of Mythos into the sub‑continent’s cyber‑security ecosystem. Participants will receive API keys and documentation that allow them to run automated scans, prioritize findings, and feed remediation workflows directly from the AI’s output.

Why it matters

The timing of Anthropic’s expansion aligns with a surge in high‑profile cyber‑attacks targeting critical‑infrastructure providers. By automating the detection of flaws that would otherwise require extensive manual review, Mythos promises to reduce the window of exposure for vulnerable systems. Early tests that revealed thousands of flaws suggest the model can uncover hidden issues in legacy codebases and newly deployed cloud services alike. For Indian organisations, the move offers a rare opportunity to tap into cutting‑edge AI security tools without building them in‑house, potentially raising the overall resilience of the country’s digital backbone. Moreover, the collaborative nature of the program—bringing together a diverse set of operators—creates a shared intelligence pool that could accelerate industry‑wide threat awareness.

The bigger picture

Anthropic’s strategy mirrors a broader trend among AI firms to commercialise specialised models for security, compliance, and risk management. Competitors such as OpenAI and Google have introduced similar offerings, but Mythos distinguishes itself with a focus on rapid vulnerability discovery rather than general‑purpose language tasks. In India, the cyber‑security market is projected to grow rapidly as businesses digitise and regulators tighten data‑protection mandates. The government’s push for a “secure by design” approach in public‑sector projects creates a receptive environment for AI‑driven tools. Anthropic’s entry therefore complements existing domestic players and may spur partnerships that blend local expertise with the model’s global threat‑intelligence capabilities.

What's next

Anthropic plans to monitor the performance of Mythos across its new user base and iterate on the model based on real‑world feedback. The company has hinted at expanding the program beyond the initial 150 organisations, potentially adding more Indian firms and expanding into other emerging markets. Stakeholders are watching for updates on integration roadmaps, pricing structures, and any regulatory guidance that may affect AI‑based security solutions. Analysts expect that as Mythos matures, its detection accuracy will improve, leading to tighter integration with security‑information‑and‑event‑management (SIEM) platforms and automated patch‑deployment tools. The rollout also raises questions about data privacy, as the model processes sensitive code; Anthropic has pledged to adhere to strict confidentiality standards, but oversight will likely intensify as adoption spreads.

Key takeaways

• Anthropic opened Mythos to about 150 organisations in over 15 countries.
• The AI model rapidly identifies security flaws, uncovering thousands of vulnerabilities in early tests.
• Indian firms and critical‑infrastructure operators are part of the new cohort.
• The move reflects a growing market for AI‑driven cyber‑security solutions globally.
• Future phases may broaden access and deepen integration with existing security tools.