people · May 13, 2026

Sophos Joins OpenAI's Trusted Access for Cyber Program for GPT-5.5 Access

Share the canonical public link.

Sophos joins OpenAI's Trusted Access for Cyber program on May 11, 2026, securing access to GPT-5.5 and a preview of GPT-5.5-Cyber for defensive workflows. The partnership enables Sophos to accelerate secure code review, vulnerability triage, malware analysis, and detection engineering using the models. Sophos integrates these tools into its agentic SOC, resolving over 50% of cases without human intervention, and enhances endpoint architecture with 60+ exploit mitigations. Previously reliant on traditional methods, Sophos now counters AI-generated zero-days through internal red teaming and operational intelligence in MDR services. OpenAI's program includes phishing-resistant authentication, achieving 89-second alert-to-response times in tests.

The body describes a specific partnership between Sophos and OpenAI for AI model access, enabling operational improvements like 50% case resolution without intervention, which justifies a 'notable' significance as a noteworthy development in cybersecurity tech integration but lacks the scale of a major strategic pivot or acquisition for higher scores. It contains multiple verifiable facts, including named companies (Sophos, OpenAI), a date (May 11, 2026), model names (GPT-5.5, GPT-5.5-Cyber), and percentages (over 50%, 60+ mitigations, 89-second times), supporting a high density score. The analysis field is explicitly '(none)', providing no strategic insight beyond the headline, warranting a zero score.

Supporting evidence