AMD Zen 5 RDSEED bug threatens cryptographic key generation

AMD has confirmed a hardware defect in the RDSEED instruction on Zen 5 processors that can return zero values for 16- and 32-bit reads, potentially weakening newly generated cryptographic keys; firmware and microcode updates are being distributed and users should apply vendor BIOS updates and consider regenerating sensitive keys.

AMD has acknowledged a hardware defect affecting the RDSEED instruction on its Zen 5 processors, logged as AMD-SB-7055 and assigned CVE-2025-62626. The company says the 16-bit and 32-bit forms of RDSEED can return zero values that may be misinterpreted as a successful entropy read, while the 64-bit form does not appear to be impacted. AMD rates the issue as high severity and emphasizes that RDSEED is a low-level source of entropy used for cryptographic keys and other security functions.

The immediate risk is that predictable or zero-valued entropy reads could produce weaker keys or nonces if software treats those returns as valid. As interim mitigations, AMD recommends preferring the 64-bit RDSEED path where possible, masking the affected RDSEED forms from software discovery, or treating a zero return as a failure and retrying. The company is distributing firmware and microcode updates to original equipment manufacturers to correct the issue.

AMD’s advisory lists product-family target dates for fixes, including AGESA TurinPI 1.0.0.8 for EPYC 9005 with a target date of November 14, 2025, and target timelines for consumer families in late November 2025. Users are advised to install vendor BIOS and microcode updates when they become available and to consider regenerating any sensitive keys created on unpatched systems. Until updates are applied, following the suggested software workarounds can reduce exposure.

66

Impact Score

Finance officials raise banking security concerns over Anthropic’s mythos model

Anthropic’s Claude Mythos has prompted urgent discussions among finance ministers, central bankers and banks over the risk that advanced cyber capabilities could expose weaknesses in critical financial systems. Governments and financial institutions are being given early access to test and strengthen defences before any broader release.

Uk delays Artificial Intelligence copyright reform

The UK government has postponed immediate copyright reform for Artificial Intelligence, leaving developers, creatives, and rightsholders to operate under existing law. Licensing, transparency, digital replicas, and future litigation are now set to shape the next phase of policy.

Memory architecture is central to autonomous llm agents

Memory design, not just model choice, determines whether autonomous agents can sustain context, learn from experience, and stay reliable over time. A practical framework centers on how information is written, managed, and read across multiple memory types.

Contact Us

Got questions? Use the form to contact us.

Contact Form

Clicking next sends a verification code to your email. After verifying, you can enter your message.