OPAQUE and TII Launch Open Standard for Verifiable AI Governance
OPAQUE 3.0 combines confidential computing and post-quantum cryptography to help organizations independently verify how AI systems operate.
News
- OPAQUE and TII Launch Open Standard for Verifiable AI Governance
- e& UAE Launches OneWork, a Unified Communications Platform Hosted Entirely in the UAE
- Alibaba Used 29 Million Fake Exchanges to Extract Claude's Intelligence, Claims Anthropic
- Anthropic's Mythos Finds Critical Flaws in Classified US Systems: Report
- Qatar Launches AI-Focused Scholarship Program to Build Future Digital Workforce
- China's LineShine Tops TOP500, Becomes World's Fastest Supercomputer
As enterprises increasingly deploy autonomous AI systems across critical infrastructure, governments are confronting a new challenge of proving that these systems behave as intended rather than relying on vendor assurances. This shift is driving demand for technologies that can independently verify how AI operates, particularly in sectors handling sensitive financial, healthcare, and government data.
At the Confidential Computing Summit in San Francisco, AI stack company OPAQUE unveiled OPAQUE 3.0, a confidential AI platform that generates cryptographic proof of how AI systems make decisions and execute tasks. The release introduces an open standard designed to create tamper-resistant audit records across the entire AI lifecycle, allowing regulators, auditors, and enterprises to independently verify that governance policies are being enforced.
The Technology Innovation Institute (TII), the applied research arm of Abu Dhabi’s Advanced Technology Research Council (ATRC), is a founding partner behind the standard and contributed to the post-quantum cryptography securing the platform.
As AI agents move beyond assisting employees to autonomously executing workflows across enterprise infrastructure, organizations are demanding stronger guarantees around compliance, accountability, and security. Traditional governance frameworks largely depend on provider claims or periodic audits. OPAQUE 3.0’s approach, in contrast, produces cryptographic evidence that every AI action complied with predefined rules enforced directly in hardware.
For governments, this capability has become increasingly linked to digital sovereignty. Rather than depending on external providers to certify the behavior of AI systems, countries can independently verify how models are trained, deployed, and operated while maintaining control over sensitive data.
The timing is important, as earlier this week the United States issued executive orders directing federal agencies to accelerate migration to post-quantum cryptography while strengthening international cooperation on quantum security. The move reflects growing recognition that quantum computing could eventually render today’s encryption standards obsolete, making long-term protection of AI systems and sensitive data a national security priority.
For the UAE, whose AI partnership with the United States expanded under the US-UAE AI Acceleration Partnership signed in 2025, the development aligns closely with the country’s own strategy. The UAE’s National Encryption Policy mandates a transition to post-quantum cryptography and identifies AI systems as critical infrastructure that require quantum-readiness assessments.
OPAQUE 3.0 combines confidential computing with post-quantum cryptography to secure AI throughout its lifecycle—from model training and fine-tuning to inference and autonomous AI agents. According to the company, it is the first platform to integrate hardware-based attestation with cryptographic protections designed to withstand future quantum attacks across the full AI stack.
For TII, the partnership extends the UAE’s growing influence in the global AI security ecosystem. Earlier this year, cryptographic AI technology developed by the institute was acquired by OPAQUE, marking one of the first large-scale commercial deployments of UAE-developed confidential AI technology by a major US technology company.
“Sovereignty in the age of AI is defined by the ability to verify, not trust. Open, transparent standards give organizations and nations the confidence to independently validate how AI systems are governed, while post-quantum cryptography preserves the confidence against the security challenges of the future,” said Dr. Najwa Aaraj, CEO of TII.
OPAQUE 3.0 will become generally available in July 2026, with additional founding partners expected to be announced alongside the standard’s broader rollout.
