Securing AI, Data, and Infrastructure at Government Scale with Steve Orrin

Steve explains his role at Intel Federal as a three part function. He helps government agencies adopt the right technologies for their missions, translates those requirements back to Intel’s internal product and engineering teams, and supports innovation where standard commercial solutions do not fully meet government needs. This two way translation ensures that future platforms align with real world mission and security demands.

The discussion moves into AI and cybersecurity, which Steve frames across three dimensions. Organizations must secure AI systems themselves, use AI responsibly to improve cybersecurity operations, and defend against adversaries that are also leveraging AI. He emphasizes that AI cannot be treated like traditional software. It requires governance, validation, and continuous monitoring across data sourcing, training, tuning, and deployment.

Steve outlines where AI is delivering tangible value today. Rather than detecting entirely new threats in isolation, AI excels at automating repetitive, high volume security tasks. By reducing the operational burden of routine alerts, patching, and triage, AI allows security teams to focus their expertise on higher impact risks and emerging threats.

A key segment of the conversation focuses on confidential computing. Steve explains how protecting data in use closes a long standing security gap that encryption at rest and in transit cannot address. Through trusted execution environments, memory encryption, isolation, and attestation, organizations can protect sensitive workloads even from compromised operating systems or untrusted cloud environments. This capability is especially relevant for AI models, intellectual property, and mission critical workloads deployed across cloud, edge, and disconnected environments.

The episode concludes with a forward looking discussion on post quantum cryptography and secure mission platforms. Steve explains why the threat is not limited to future quantum computers, but to data being harvested and stored today for later decryption. Government driven timelines are accelerating adoption, and commercial industries will benefit from following the same path as compliant products become broadly available.

Resources mentioned in this episode