Key Takeaways
- GPT-5.4-Cyber has been deployed to a limited group of authenticated cybersecurity experts
- The rollout follows Anthropic’s Mythos security AI by exactly seven days
- U.S. Treasury Secretary Scott Bessent and Fed Chair Jerome Powell flagged Anthropic’s Mythos as a potential risk
- OpenAI intends to scale access from a few hundred to several thousand vetted professionals soon
- Ramp’s data reveals nearly 33% of American companies subscribed to Anthropic services in March
OpenAI has introduced a specialized artificial intelligence system designed exclusively for cybersecurity applications. Named GPT-5.4-Cyber, this model is currently accessible to a carefully vetted cohort of security professionals.
The deployment occurred on Tuesday, April 15, arriving exactly seven days after Anthropic unveiled its security-oriented system, Mythos, through a restricted release.
GPT-5.4-Cyber represents what OpenAI characterizes as a “more permissive” iteration of its GPT-5.4 platform. The system features reduced restriction thresholds for cybersecurity operations that the company classifies as authorized protective activities.
Currently, usage is restricted to members of OpenAI’s Trusted Access for Cyber initiative. This program debuted in February and mandates rigorous identity authentication for users seeking model privileges.
The company announced plans to expand participation from several hundred early testers to thousands of authenticated security professionals over the next several weeks. OpenAI emphasized that identity verification will remain mandatory for all prospective users.
OpenAI initiated the integration of cybersecurity-specific protective measures into its systems during 2025. In March, the company unveiled Codex Security, a tool designed to detect and remediate software security flaws.
Regulatory Scrutiny of Anthropic’s Mythos
Anthropic’s Mythos platform has quickly attracted scrutiny from senior government officials. During the previous week, Treasury Secretary Scott Bessent alongside Federal Reserve Chair Jerome Powell cautioned financial industry executives to carefully consider the threats posed by Mythos, Bloomberg reported.
Anthropic made the strategic decision to distribute Mythos to approximately 40 major corporations to strengthen their defensive security capabilities. The restricted distribution reflects the company’s apprehension about unauthorized access to the technology.
The director of Ireland’s National Cyber Security Centre testified before a parliamentary committee on April 14, predicting that malicious actors will likely develop comparable capabilities to Mythos before year’s end.
Anthropic co-founder Jack Clark projected that competing organizations will unveil similar technologies within the coming months. He further anticipated that open-weight Chinese models featuring equivalent capabilities might emerge within 12 to 18 months.
OpenAI’s Enterprise Market Strategy
OpenAI has intensified its emphasis on corporate clients, a segment where Anthropic has rapidly established significant presence. According to analytics from payment processor Ramp, approximately 33% of American enterprises purchased Anthropic solutions during March.
OpenAI has simultaneously reduced investment in lower-margin initiatives. The organization has suspended development of an adult-oriented ChatGPT variant and temporarily halted its Stargate UK initiative.
Microsoft, serving as OpenAI’s primary financial supporter, has committed to leasing data center infrastructure at a location initially designated for the Stargate Norway development. A separate agreement with British AI infrastructure company Nscale similarly collapsed.
OpenAI revealed plans to establish its inaugural permanent London headquarters in 2027, featuring capacity for over 500 employees. The company identified London as its most substantial research facility beyond American borders.
OpenAI has additionally unveiled a new “superapp” concept influenced by Anthropic’s approach, and introduced specialized healthcare AI capabilities shortly following Anthropic’s deployment of health-centered Claude functionality.
