OpenAI Unveils GPT-5.4-Cyber Model and Expands Cybersecurity Program Following Anthropic's Vulnerability Discovery
What happened
On April 14, 2026, OpenAI released GPT-5.4-Cyber, a new AI model focused on cybersecurity, alongside an expanded Trusted Access for Cyber program.
The Context
- OpenAI's response: This launch follows Anthropic's Claude Mythos Preview, which revealed thousands of high-severity vulnerabilities, raising concerns about AI's potential misuse.
- TAC program expansion: OpenAI's Trusted Access for Cyber program now includes tiered access for verified cybersecurity professionals, enhancing defensive capabilities.
- Industry dynamics: The contrasting approaches of OpenAI and Anthropic highlight a growing competition in the AI-cybersecurity landscape, with differing philosophies on access and control.
The Number
of high-severity vulnerabilities were discovered by Anthropic's Claude Mythos Preview, underscoring the urgent need for robust cybersecurity measures.
Takeaway
Expect increased competition and innovation in AI-driven cybersecurity solutions as companies respond to evolving threats.
Tech, science, and startup news including AI.
"Irish tech outlet covering innovation and AI."
— A47 Editor
After Anthropic, OpenAI launches cyber-specific AI model
OpenAI has launched GPT-5.4-Cyber, a specialized variant of its GPT-5.4 model, aimed at enhancing defensive cybersecurity capabilities. This model lowers the refusal boundary for what constitutes 'legitimate' cybersecurity work, allowing for broader ...
Tech industry coverage with AI angles.
"Mainstream tech news intersecting with AI policy and culture."
— A47 Editor
Like Anthropic, OpenAI Will Share Latest Technology Only With Trusted Companies
OpenAI has announced the limited release of its new technology, GPT-5.4-Cyber, which is specifically designed to identify security vulnerabilities in software. This model represents a significant advancement in OpenAI's efforts to enhance cybersecuri...
Tech policy, trends, and innovation news.
"The New York Times is a globally recognized newspaper offering authoritative reporting with a center-left editorial stance."
— A47 Editor
Like Anthropic, OpenAI Will Share Latest Technology Only With Trusted Companies
OpenAI has announced the limited release of its new technology, GPT-5.4-Cyber, which is specifically designed to identify security vulnerabilities in software. This model represents a significant advancement in OpenAI's efforts to enhance cybersecuri...
Latest tech news, product reviews, and analysis for consumers and professionals.
"CNET delivers accessible and detailed technology reporting, including trusted product reviews and how-to guides."
— A47 Editor
OpenAI Has a New GPT-5.4-Cyber Model. Here's Why You Can't Use It
OpenAI has introduced its new GPT-5.4-Cyber model, a significant enhancement of its Trusted Access for Cyber program, which appears to be a strategic response to Anthropic's Project Glasswing aimed at preventing AI-driven cyberattacks.
Market-moving headlines impacting equities, bonds, and related risk assets.
"Real-time catalysts and volatility drivers across indices and sectors."
— A47 Editor
OpenAI unveils GPT-5.4-Cyber a week after rival’s announcement of AI model
OpenAI has introduced GPT-5.4-Cyber, a new AI model, just a week after a competitor announced its own AI offering. This launch reflects OpenAI's ongoing commitment to advancing artificial intelligence technology and maintaining its competitive edge i...
Emerging technologies, digital transformation, IT, and cultural impact of tech.
"WIRED covers the intersection of technology, culture, and politics with a progressive, forward-looking editorial stance."
— A47 Editor
In the Wake of Anthropic’s Mythos, OpenAI Has a New Cybersecurity Model—and Strategy
OpenAI has unveiled its new cybersecurity-focused model, GPT-5.4-Cyber, which aims to enhance safeguards against cyber threats. The company asserts that its existing measures sufficiently reduce cyber risks while this new model represents a strategic...