ODINI examines a containment layer stronger than an ordinary air gap: a computer placed behind metal shielding intended to suppress electromagnetic leakage. Malware varies the load on CPU cores in a controlled pattern. Changes in current through the processor and its power-delivery path alter the low-frequency magnetic field around the machine, allowing…
Inspect the ClaimsCase · DiggingBeagle record
ODINI exfiltrated through CPU-generated magnetic fields despite Faraday shielding
ODINI research modulated low-frequency magnetic fields by varying CPU load and demonstrated a receiver path intended to work through metal shielding that blocks higher-frequency electromagnetic emissions.
Scope
Research demonstration requiring malicious code on the protected computer and a nearby magnetic receiver.
Assessment profile
Assessment method- Setting
- research demonstration
- Exploitation
- demonstrated
- Evidence
- Grade B
Basis and provenance
The CPU-load magnetic channel and shielded/VM prototype claims describe a research demonstration, with no demonstrated deployment prevalence. Assessment is a desk review of the retained canonical Claims and cited Source metadata at their recorded cutoff, not a new external verification. Impact remains ungraded.
Assessed Sep 24, 2026 using diggingbeagle.assessment/1.
Evidence basis: CLM-ODINI-MECHANISM · CLM-ODINI-SHIELDING · ODINI: Escaping Sensitive Data from Faraday-Caged, Air-Gapped Computers via Magnetic Fields
At a glance
ODINI's demonstrated impact is a local magnetic exfiltration path after compromise. Its broader implication is architectural: 'inside a Faraday cage' is not a complete statement of physical isolation unless the enclosure is characterized for the frequencies and field types an attacker can induce. The result should not be generalized to universal success…
Read the implicationsThe cited work does not establish universal success across Faraday cages, rack layouts, processor generations or facility-scale shielding. Practical range and signal quality depend on hardware, receiver sensitivity, placement and environmental magnetic noise. No cited evidence establishes autonomous AI use of ODINI.
Limits and uncertaintyFull account
ODINI examines a containment layer stronger than an ordinary air gap: a computer placed behind metal shielding intended to suppress electromagnetic leakage. Malware varies the load on CPU cores in a controlled pattern. Changes in current through the processor and its power-delivery path alter the low-frequency magnetic field around the machine, allowing data to be encoded in that field.
The physical distinction matters. A conventional Faraday enclosure is effective against many electric-field and higher-frequency electromagnetic emissions, but low-frequency magnetic fields can penetrate metal shielding differently. The researchers used a nearby magnetic receiver to detect the modulated field and reported that their prototype did not require special privileges and could operate from an isolated virtual machine under their tested conditions.
ODINI does not prove that every Faraday-caged system leaks usable data. Receiver placement, field strength, shielding design, workload noise and hardware topology remain decisive. It demonstrates that shielding has to be designed for the relevant physical field rather than treated as a universal communications barrier.
Mechanism and trust boundary
- 01
Run code inside the protected machine
The channel assumes attacker-controlled code is already executing, including in the researchers' tested VM scenario.
Boundary: shielded host / attacker-controlled workload
- 02
Modulate CPU-core load
The malware varies processor utilization according to the data it wants to transmit.
Boundary: secret data / electrical workload
- 03
Generate a low-frequency magnetic pattern
Changing processor current produces corresponding changes in the surrounding magnetic field.
Boundary: electrical activity / magnetic field
- 04
Cross the shielding boundary
The low-frequency magnetic component can propagate through shielding that is effective against other electromagnetic emissions.
Boundary: shielded enclosure / external physical space
- 05
Measure with a nearby magnetic sensor
A prepared receiver observes the field changes and decodes the transmitted symbols.
Boundary: magnetic field / recovered data
Timeline
- Feb 8, 2018Event type unspecified
ODINI preprint released
The researchers published the magnetic covert-channel design and prototype results.
- Aug 29, 2019Event type unspecified
Peer-reviewed publication metadata recorded
The corresponding IEEE Transactions publication record was added for the peer-reviewed version.
Claims & evidence
CLM-ODINI-MECHANISMODINI demonstrated data modulation over low-frequency magnetic fields generated by controlled CPU-core load on an air-gapped computer.supported
Basis: reported finding
- supportsODINI: Escaping Sensitive Data from Faraday-Caged, Air-Gapped Computers via Magnetic Fieldsrole not specified
Abstract
CLM-ODINI-SHIELDINGThe authors designed the channel to penetrate metal shielding such as Faraday cages and report operation without special privileges, including from isolated virtual machines in their prototype conditions.supported
Basis: reported finding
- supportsODINI: Escaping Sensitive Data from Faraday-Caged, Air-Gapped Computers via Magnetic Fieldsrole not specified
Abstract, privilege and VM statements
Implications
ODINI's demonstrated impact is a local magnetic exfiltration path after compromise. Its broader implication is architectural: 'inside a Faraday cage' is not a complete statement of physical isolation unless the enclosure is characterized for the frequencies and field types an attacker can induce. The result should not be generalized to universal success through arbitrary shielded facilities.
Controls and mitigations
- Use physical exclusion zones that prevent untrusted magnetic sensors from being positioned close enough to protected equipment.
- Characterize shielding against relevant low-frequency magnetic fields rather than assuming high-frequency RF attenuation is sufficient.
- Monitor suspicious patterned CPU workloads where the assurance level justifies side-channel detection.
Unknowns and contradictions
- The cited work does not establish universal success across Faraday cages, rack layouts, processor generations or facility-scale shielding.
- Practical range and signal quality depend on hardware, receiver sensitivity, placement and environmental magnetic noise.
- No cited evidence establishes autonomous AI use of ODINI.
Sources and citation
Material revision history
- Sep 19, 2026 · Published version · first publication · revision 15
- Sep 19, 2026 · Published version · material revision · revision 20Public fields changed: attack_path, implications, mitigations, narrative, timeline, unknowns
- Sep 25, 2026 · Published version · material revision · revision 36Public fields changed: assessment_profile