OpenAI has offered to open access to its cybersecurity model as part of its EU Cyber Action Plan, which aims to democratize defensive AI tools for European institutions and businesses. This offer comes as the European Commission is engaging with major AI developers to assess various cybersecurity models in light of growing concerns over advanced AI capabilities. Meanwhile, Anthropic has not yet committed to similar access for its Mythos model, indicating a more cautious approach to sharing its AI technology.

OpenAI: OpenAI is an artificial intelligence research and deployment company best known for developing the GPT series of large language models used for general-purpose and specialized tasks. It recently introduced GPT-5.5-Cyber, a model focused on cybersecurity applications like vulnerability identification and defensive strategies, as part of its Trusted Access for Cyber program. In this news, OpenAI has offered the European Union access to this model via the OpenAI EU Cyber Action Plan to collaborate with policymakers and businesses on enhancing cyber resilience.
Anthropic: Anthropic is an AI safety-focused company that develops the Claude family of large language models with an emphasis on reliability and risk mitigation. It recently unveiled Claude Mythos, an advanced model exhibiting strong capabilities in cybersecurity tasks such as organizing complex cyber operations. The news highlights that Anthropic has not yet agreed to provide the EU with access to Mythos, with discussions remaining at an earlier stage compared to OpenAI.

`json
{
“OpenAI Offer”: “OpenAI has proposed to provide access to its cybersecurity model for European institutions and businesses.”,
“EU Engagement”: “The European Commission is engaging with major AI developers to assess cybersecurity models amid growing concerns about the capabilities of advanced AI.”,
“Model Differentiation”: “While OpenAI is ready to offer access to its tools, Anthropic has not yet committed to sharing its cybersecurity model.”
}
`