Exclusive: Anthropic left details of an unreleased model, an upcoming exclusive CEO event, in a public database

Anthropic, the AI safety-focused company behind Claude, suffered a significant security lapse by leaving details of an unreleased AI model and an exclusive CEO event in a publicly accessible content management system (CMS). According to a report by Fortune, cybersecurity researcher Alexandre Pauwels from the University of Cambridge discovered close to 3,000 unpublished assets—including images, PDFs, and blog post drafts—linked to Anthropic's website that were not meant for public view. The cache contained specifics about a forthcoming AI model, which could include performance benchmarks or architectural details, and information about a planned high-level executive event. The company took steps to secure the data only after being notified by Fortune on Thursday, highlighting a reactive rather than proactive security posture.

This incident represents more than a simple data exposure; it's a strategic leak that could impact Anthropic's competitive positioning in the crowded AI market. With companies like OpenAI, Google, and Meta aggressively announcing new models, controlling the narrative around product launches is crucial. The leaked details about an exclusive CEO event suggest Anthropic was preparing a coordinated reveal to key stakeholders and potentially enterprise clients. Such security failures not only risk intellectual property but also undermine trust with partners who expect rigorous data handling from a company whose brand is built on safety and responsible development. The fact that basic CMS permissions allowed this breach indicates potential gaps in Anthropic's internal security protocols despite their focus on long-term AI safety.