cybersecurity ai-safety openai trusted-access

Trusted access for the next era of cyber defense

openai.com

April 14, 2026

7 min read

🔥🔥🔥🔥🔥

49/100

Summary

The Trusted Access for Cyber (TAC) program is expanding to include thousands of verified individual defenders and hundreds of teams focused on protecting critical software. The program emphasizes democratized access, iterative deployment, and ecosystem resilience while fine-tuning models for defensive cybersecurity use cases in anticipation of more advanced models from OpenAI.

Key Takeaways

OpenAI is scaling its Trusted Access for Cyber (TAC) program to thousands of individual defenders and hundreds of teams to enhance cybersecurity efforts.
The new GPT-5.4-Cyber model is specifically fine-tuned for defensive cybersecurity use cases.
OpenAI's cybersecurity strategy focuses on democratized access, iterative deployment, and investing in ecosystem resilience to support defenders against cyber threats.
The company has launched Codex Security to help identify and fix vulnerabilities at scale, enhancing the capabilities of cybersecurity defenders.

Read original article

View story →

Coverage & discussions

Hacker News · openai.com

Trusted access for the next era of cyber defense ↗

Community discussions

Hacker News discussion ↗

Historical count: 5 · 5 saved in sample · Historical sample

Analysis ready

Community Sentiment

Mixed

Positives

The introduction of a Security button in the Codex section indicates a proactive approach to integrating AI with cybersecurity, allowing users to run vulnerability scans on their GitHub repositories.
The verification process for 'Trusted Access' suggests a commitment to enhancing security measures in AI applications, which is crucial for building trust in AI tools.

Concerns

Many users express frustration that the 'Trusted Access' verification has not unlocked any new features in the OpenAI API or Codex models, indicating a lack of transparency or functionality.
There are concerns that the focus on enterprise and government markets may lead to neglecting consumer needs, which could stifle innovation and accessibility in AI technologies.