Researchers have disclosed the DDRop attack, a hardware vulnerability targeting Intel TDX and AMD SEV-SNP confidential computing environments. Disclosed on September 14, 2026, the exploit utilizes a custom interposer costing under $200 to silently drop memory writes on DDR5 memory, bypassing freshness checks and compromising virtual machine integrity across major cloud providers.
The Hardware Root of the DDRop Vulnerability
Confidential computing relies on Trusted Execution Environments (TEEs) to protect data in use by encrypting memory at the hardware level. However, modern cloud servers handle massive volumes of data, forcing hardware designers to make architectural trade-offs. To cover the large amount of memory that a cloud server uses, designs for Intel Trust Domain Extension (TDX) and AMD Secure Encrypted Virtualization-Secure Nested Paging (SEV-SNP) omit a guarantee known as freshness.
Freshness ensures that the processor can confirm that data read from memory is the most recently written version. Without it, the system remains blind to certain physical interventions. Enter DDRop. According to findings from researchers at KU Leuven, ETH Zurich, Durham University, and Google, attackers can exploit this architectural blind spot by deploying a physical interposer directly onto the memory bus.
This isn’t a passive packet sniffer like TEE.fail attacks. DDRop is an active interposer attack capable of operating at full DDR5 speeds. By silently dropping memory writes, the hardware modification allows malicious actors to roll back memory states, bypass encryption layers, and achieve full virtual machine control on Intel TDX. This includes reading private memory and successfully forging attestations.
Infrastructure Impact Across Major Cloud Providers
Because the vulnerability is rooted in physical hardware design rather than a software bug, no simple patch exists. Major cloud server environments—including infrastructure run by AWS, Microsoft Azure, and Google Cloud—are exposed to the underlying architectural risk.
The attack vector relies on an inexpensive piece of hardware—an interposer priced at under $200—sitting between the memory module and the motherboard slot. This opens up vectors for supply chain tampering or rogue employees with physical access to enterprise data centers.
Mitigating the DDRop threat requires a shift in how data centers approach physical security and boot integrity. Security teams running confidential workloads are currently being urged to evaluate Intel’s cryptographic-integrity mode where available, alongside implementing boot-time interposer detection mechanisms to catch unauthorized hardware modifications before hypervisors spin up.
The 30-Second Verdict for Enterprise Security Teams
- The Threat: A sub-$200 hardware interposer that silently drops DDR5 writes.
- The Targets: Intel TDX and AMD SEV-SNP confidential computing architectures.
- The Impact: Full VM control, memory inspection, and forged attestations.
- The Fix: No simple software patch exists; relies on hardware-level mitigations and boot-time detection.
The rush to scale cloud memory capacities outpaced the hardware-level freshness guarantees required to keep modern TEEs secure.
