AI speeds software development

Artificial intelligence is significantly accelerating the pace of software development. However, this progress is also introducing new cybersecurity challenges as systems become more complex and interconnected.

Oracle and OpenAI collaborate

Oracle and OpenAI are working together to provide businesses with an integrated solution to address the growing threats posed by autonomous AI agents and the increasing number of vulnerabilities in information systems.

Security concerns highlighted

At the Oracle AI World 2026 conference in Las Vegas, Sam Altman, CEO of OpenAI, stressed the need to understand and mitigate security risks emerging from AI. He emphasized that these risks must be actively managed to prevent exploitation.

AI detects hidden flaws

Clay Magouyrk, CEO of Oracle, noted that AI not only introduces new risks but also helps uncover vulnerabilities that traditional tools have often missed. These hidden flaws can now be identified and addressed more effectively.

New security solution introduced

Oracle unveiled Oracle AI Security Mesh, a solution that merges OpenAI's AI models with Oracle's security testing tools. This system is designed to identify, manage, and fix vulnerabilities across various environments.

Solution covers multiple stages

The solution spans several stages, from identifying vulnerabilities to proposing and managing corrective actions. An AI agent can analyze potential attack paths, assess the risk of exploitation, and recommend appropriate fixes.

Solution works across environments

Oracle AI Security Mesh is not limited to Oracle Cloud Infrastructure. It is also designed to work with third-party cloud infrastructures and on-premise systems used by businesses.

Customer validation required

During a demonstration, Oracle showed how an AI agent could detect vulnerabilities in unsecured APIs and suggest changes to access parameters. However, these changes are not implemented automatically; customers must approve them before any adjustments are made.

Managing AI agents is a priority

Beyond detecting vulnerabilities, Oracle and OpenAI are addressing the growing number of AI agents interacting with business applications, data, and infrastructure. These agents offer new automation possibilities but also increase the complexity of cybersecurity.

Cybersecurity strategy evolves

Managing permissions, securing APIs, and controlling autonomous systems have become central to modern cybersecurity strategies. Sam Altman noted that AI is now capable of detecting a wide range of vulnerabilities, requiring new approaches to security.

Shift in cybersecurity focus

The approach reflects a broader shift in cybersecurity from detection and response to a continuous process that integrates the identification, verification, and correction of vulnerabilities.

Beyond security: AI integration

The partnership between Oracle and OpenAI extends beyond cybersecurity. They also introduced Oracle AI Workspace, an environment that connects ChatGPT with Oracle-managed enterprise data and applications.

Improving user efficiency

With Oracle AI Workspace, users can access professional resources directly through ChatGPT, based on their permissions. This aims to streamline business tasks by reducing the need to switch between multiple applications.