☠️
EXFILTRATING
DATA...
01010111 10110010 00101001 11000110 01101100 10011011 00010001 11100101 01000010 10101111
[ <-- BACK_TO_SEARCH ]
[ VENDOR STATS ]
SOON 🐛 CWE-229
👁️
Common Weakness Enumeration - Identifies the specific category of software weakness or flaw.

CVE-2025-35973

CVE-2025-35973 Vulnerability in Intel
📅 PUBLISHED: 2026-08-11 00:00 UTC 🔄 UPDATED: 2026-09-24 06:00 UTC
[ PATCH UNDER REVIEW ↓ ] Remains UNDER REVIEW until an independent source confirms the fix
[ DEFENSE PATCH ↓ ]
💣 PoC: Public
👁️
Proof of Concept - Indicates if functional exploit code is publicly available, drastically increasing the risk of immediate attacks.
🎯 EPSS: 0.10%
👁️
0.10% probability
2.6× above the median CVE
Tier: ELEVATED
30-day exploitation window · FIRST.org · updated daily
🥷 MITRE: T1068
👁️
MITRE ATT&CK Tactic - Describes the strategy hackers use to execute this attack.
VULNERABLE: version not auto-detected — verify your deployed version locally
CVSS SEVERITY SCALE
9.0–10.0 CRITICAL
7.0–8.9 HIGH
4.0–6.9 MEDIUM
0.1–3.9 LOW
0.0 NONE
SOON
PENDING NVD
■ tracking this CVE ✓ independent patch confirmation ⧗ same-origin mirror (GHSA) ⚠ active exploitation verification policy →

Executive Summary

CVE-2025-35973 is a privilege-escalation vulnerability in some Intel processors caused by improper handling of values during Ring 0 execution contexts (kernel, hypervisor, and bare-metal operating systems). Successful exploitation requires an authorized adversary who already holds a privileged user role, a high-complexity attack procedure, local access, and special internal knowledge of the target platform, with no user interaction required. If exploited, the vulnerability enables broad confidentiality and integrity compromise of the entire system while leaving availability unaffected; no public exploit, patch, or CISA KEV listing exists as of this reporting date.

⚡ AUTOMATE YOUR CVE INTELLIGENCE

This record — and 170,000+ others — are available via the ValtersIT API. One endpoint replaces five: NVD + GitHub CVE + CISA KEV + EPSS + Nuclei + Exploit-DB, unified.

›Patch confidence — UNVERIFIED to CONFIRMED, cross-referenced NVD + RHSA + Debian
›Exploit & PoC — poc_status, poc_url, has_exploit. Not in NVD.
›Shodan dorks — ready-made queries to find exposed vulnerable systems
›Detection rules — YARA · Wazuh · Elastic (all plans) · Sigma (Pro)
›pending_fields — know exactly which of 7 fields NVD hasn't processed yet
›Free backfill — data incomplete when you paid? Re-check within 30 days = 0 credits
PAYG €0.08/lookup Standard €19/mo (was €29) · 5k credits Pro €65/mo (was €99) · 20k credits FOUNDING PRICE · LOCKED FOR LIFE · ENDS 31 DEC 2026
[ FROM €19/MO ]
Free sandbox key available
No signup required

Technical Analysis

The NVD description characterizes this flaw through CWE-229 (Improper Handling of Values). In processor design, this defect class indicates that certain value inputs passed to architectural or microarchitectural logic are not correctly validated or constrained before being used in privileged operations. The affected execution domains — Ring 0 (kernel), hypervisor, and bare-metal OS — are the highest-privilege software layers on an x86 platform, meaning the improper value handling occurs in a context where any logic error can directly affect the processor's privilege separation boundaries.

The attack prerequisites are unusually restrictive and indicate a non-trivial exploitation path. The adversary must already be an authorized user with a privileged account, possess special internal knowledge of the vulnerable system (suggesting undocumented or model-specific processor behavior is involved), and conduct a high-complexity attack. The attack is strictly local and requires no user interaction. The impact profile is telling: the directly vulnerable component suffers only low confidentiality and low integrity impact, but the subsequent system-level impact is rated high for confidentiality and high for integrity, with availability unaffected. This two-tier impact structure is consistent with a processor-level flaw in which a limited register/value corruption in Ring 0 translates into a full break of kernel, hypervisor, or bare-metal memory isolation.

It must be emphasized that the NVD entry provides no CNA title, no vendor advisory reference, no affected version list, and no technical detail on the specific instruction sequence, register set, or microarchitectural mechanism involved. Whether this is a speculative-execution side channel, a fault-injection target, or an architectural value-handling bug cannot be confirmed from the available data. The report therefore treats the vulnerability strictly at the class level: a Ring 0 privilege-escalation defect in Intel processor value handling, triggerable only by a local, highly privileged, well-informed attacker.

Exploitation Scenarios

No public proof-of-concept or exploit code exists for CVE-2025-35973 (Nuclei template: absent; exploit code: absent; PoC status: unknown). The attack chain below is therefore inferred from the vulnerability class and the stated requirements in the NVD description.

  1. Reconnaissance and target identification. An attacker who already holds a privileged user account on the target system (or has compromised one) first enumerates the platform profile: exact Intel CPU model and stepping, installed microcode version, firmware/BIOS revision, and whether the target runs a bare-metal OS, a hypervisor, or a kernel with ring-0 attack surface exposed. This reconnaissance is entirely local (e.g., CPUID-style queries, firmware version reads) and matches the "special internal knowledge" requirement — the attacker must match the crafted values to the specific processor implementation.

  2. Triggering the flaw. Leveraging CWE-229, the attacker delivers specially crafted values to the processor's vulnerable handling logic through legitimate privileged interfaces. Because the attacker is already privileged, the exploitation does not require breaking an initial boundary — instead, it uses an existing ring-0-capable channel to feed malformed or out-of-contract values that the processor fails to handle correctly. The high-complexity rating and the absence of user interaction suggest a deterministic but difficult sequence of operations against specific processor microarchitectures.

  3. Payload and execution method. At the vulnerability-class level, the improper handling of values causes the processor to perform a privileged operation with incorrect assumptions, allowing the attacker's controlled context to gain access rights beyond the intended privilege boundary. This would manifest as kernel-mode, hypervisor-mode, or bare-metal code execution with memory read/write capabilities across the entire system, including memory regions that should be isolated from the attacker's current privilege domain. No specific payload, function name, or instruction sequence can be described because none is publicly documented.

  4. Post-exploitation impact. Consistent with the NVD impact data (subsequent confidentiality: high, integrity: high, availability: none), a successful exploit yields full read/write access to the OS kernel and, in hosted environments, to the hypervisor and co-resident virtual machines. The attacker could extract credentials, subvert security products, implant persistent firmware/kernel-level backdoors, or break out of guest isolation on shared virtualization infrastructure. The absence of availability impact suggests the attack does not crash the system, which aids covert operation; however, the attacker must already hold a privileged position, so the marginal escalation is from privileged user to full system compromise.

Impact Assessment

No CVSS vector or score is published for CVE-2025-35973, so a quantitative severity figure cannot be provided. Qualitatively, the vulnerability is severe in outcome (system-level confidentiality and integrity both rated high) but constrained in reachability: it requires a pre-existing privileged user, high attack complexity, and special internal knowledge, all with local access only. The low EPSS score (0.00102, at the 1.1 percentile of exploited CVEs) and the absence of any exploit-in-the-wild evidence or CISA KEV inclusion further indicate that active exploitation is currently rare or nonexistent.

The greatest business risk applies to organizations operating shared infrastructure — cloud providers, virtualized data centers, and multi-tenant hosting platforms — where a single privileged user on one guest could theoretically escalate to the hypervisor and compromise adjacent tenants. Enterprises with large fleets of Intel-based servers or workstations should also consider that the vulnerability is currently unpatched (patch confidence 0/4), meaning mitigating controls must be procedural while awaiting Intel's disclosure and microcode/firmware remediation. The confidentiality and integrity of the affected system are the blast radius; availability is explicitly not impacted, so business continuity is not at risk from this CVE.

Detection Guidance

Because CVE-2025-35973 is a local, high-complexity processor-level flaw with no public exploit artifact, pure network detection is of limited value. The rules below therefore focus on two observable behaviors: (a) the delivery and installation of kernel-mode payloads (the likely vehicle for ring-0 code execution once the value-handling flaw is triggered) and (b) anomalous privileged activity following a suspicious driver load. These are generic behavioral controls designed to be refined once Intel publishes technical details and any in-the-wild payload is observed.

Sigma Rule

title: Suspicious Kernel Driver Load – Possible Ring-0 Privilege Escalation (CVE-2025-35973)
id: 6f4e4f4e-9c4e-4f0e-8c4e-000000000000
status: experimental
description: Detects kernel driver (.sys) loads from atypical directories such as Temp, Users\Public, or ProgramData. Loading an unsigned or unusual kernel driver is a common channel for executing ring-0 code when a processor-level privilege-escalation flaw such as CVE-2025-35973 is abused from an already-privileged user context.
references:
    - https://nvd.nist.gov/vuln/detail/CVE-2025-35973
logsource:
    product: windows
    category: driver_load
detection:
    selection_temp:
        EventID: 6
        ImageLoaded|contains:
            - '\Temp\'
    selection_public:
        EventID: 6
        ImageLoaded|contains:
            - '\Users\Public\'
    selection_programdata:
        EventID: 6
        ImageLoaded|contains:
            - '\ProgramData\'
    condition: selection_temp or selection_public or selection_programdata
falsepositives:
    - Legitimate software updaters that stage drivers in Temp directories
    - Legacy enterprise installers extracting signed drivers to Public or ProgramData
level: high
tags:
    - attack.privilege_escalation
    - attack.t1068

YARA Rule

rule CVE_2025_35973_Ring0_Payload_Hunt
{
    meta:
        author = "Valtersit Threat Intelligence"
        description = "Hunts for Windows PE payloads carrying ring-0 privilege-escalation context strings, consistent with kernel/hypervisor agents that would be leveraged after abusing CVE-2025-35973. No public payload exists, so this rule is intentionally generic and should be treated as an anomaly signal."
        date = "2025-06-19"
        cve = "CVE-2025-35973"
        reference = "https://nvd.nist.gov/vuln/detail/CVE-2025-35973"
    strings:
        $pe = { 50 45 00 00 }                    // "PE\0\0"
        $s1 = "Ring 0" ascii wide nocase
        $s2 = "Hypervisor" ascii wide nocase
        $s3 = "Bare Metal" ascii wide nocase
        $s4 = "privilege escalation" ascii wide nocase
    condition:
        uint16(0) == 0x5A4D and $pe and 2 of ($s*)
}

Suricata IDS Rule

Network detection of this local CPU vulnerability is inherently limited. The following rule detects one plausible delivery vector: a Windows kernel driver (PE executable) with a .sys extension downloaded over HTTP, which could represent the ring-0 payload stage of an exploitation chain. A high volume of false positives is unlikely but possible with legitimate driver distribution portals.

alert http any any -> any any (msg:"CVE-2025-35973: Kernel driver (.sys) download detected - possible ring-0 payload delivery"; flow:to_client,established; file.name:/\.sys$/i; file.magic:"PE32 executable"; classtype:attempted-privilege-escalation; sid:202535973; rev:1;)

Elastic Detection

sequence by host.id with maxspan=10m
  [ file where event.type == "creation" and file.extension == "sys"
      and (file.path : "C:\\Windows\\Temp\\*" or file.path : "C:\\Users\\Public\\*" or file.path : "C:\\ProgramData\\*") ]
  [ process where event.type == "start" and process.name == "sc.exe"
      and process.args : ("create*", "start*") ]

This EQL sequence correlates a kernel driver appearing in an atypical directory with a subsequent Windows service creation via sc.exe — a pattern consistent with installing a ring-0 component on a compromised host.

Splunk SPL Query

index=windows sourcetype="XmlWinEventLog:Microsoft-Windows-Sysmon/Operational" EventCode=6
| where match(ImageLoaded, "(?i)(Temp|Public|ProgramData).*\.sys")
| stats count by host, ImageLoaded, Signature, Hashes
| sort - count

The query surfaces Sysmon driver-load events (Event ID 6) where the loaded image path contains Temp, Public, or ProgramData and the file extension is .sys. Add a follow-up correlation to security events for privileged group membership changes (e.g., Event ID 4728/4732) on the same endpoints within a short time window.

Wazuh OSSEC Rule

<group name="sysmon,">
  <rule id="100973" level="12">
    <if_sid>550</if_sid>
    <field name="win.eventdata.imageLoaded" type="pcre2">(?i)(Temp|Public|ProgramData).*\.sys</field>
    <description>CVE-2025-35973: Suspicious kernel driver loaded from atypical path - possible ring-0 privilege escalation.</description>
    <group>privilege_escalation,</group>
  </rule>
</group>

This Wazuh rule inherits from the Sysmon base rule (550) and triggers at level 12 on driver-load events whose image path indicates an anomalous staging directory, flagging potential ring-0 payload installation for immediate SOC review.

Mitigation & Patch Guidance

Patch confidence for CVE-2025-35973 is 0/4 — UNPATCHED. No vendor advisory, firmware update, microcode revision, or software patch has been identified from any tracked source.

  1. Immediate mitigations. Since exploitation requires a privileged local user, restrict the population of privileged accounts and reduce local administrative surface: implement least-privilege principles, remove unnecessary local admin rights, enforce strong multi-factor authentication for privileged accounts, and monitor privileged account activity closely. Segment hypervisor management interfaces and virtualized workloads so a compromised privileged user on one tenant cannot easily reach co-resident systems. Enable virtualization-based security (VBS/HVCI) and Windows Credential Guard where the platform supports it, as these raise the cost of a ring-0 foothold. Document and track the Intel microcode revision on all affected systems so that remediation can be verified once firmware is released.
  2. Patch and upgrade steps. No patch is available today. Subscribe to Intel's Product Security Center disclosures and monitor OEM platform/firmware release notes; Intel's remediation for processor-level issues is normally delivered as CPU microcode bundled into BIOS/UEFI updates. When the Intel advisory and updated microcode are published, apply the new firmware to all affected servers, workstations, and bare-metal/hypervisor hosts, then verify the running microcode revision post-reboot. Reassess patch confidence on every vendor advisory release.
  3. Long-term hardening. Adopt a hardware lifecycle policy that prioritizes platforms with mature security features and attested boot. Where available, use hardware attestation to verify the integrity of the ring-0 software stack at boot time. Maintain fleet-wide inventory of CPU models, steppings, and microcode versions so that future Intel processor disclosures can be triaged in hours rather than weeks. Finally, treat this CVE as a reminder that privileged-user compromise is the precursor to the most severe outcomes — invest in privileged access management (PAM), behavioral analytics for kernel-level anomalies, and rapid firmware deployment capability.

References

As of this report's publication date, there are no known threat-actor campaigns, exploit-in-the-wild evidence, or CISA KEV listings for CVE-2025-35973. The vulnerability is unpatched; readers are advised to monitor Intel's advisory feed and this NVD record for updates.

👁️ SHODAN DORKS (THREAT HUNTING)

What is this? Security researchers use these exact search queries on Shodan.io (a search engine for internet-connected devices) to find vulnerable systems globally. You can copy these dorks to see if your organization's IP space is exposed.

Intel(R) Xeon
Intel(R) Core i9
hypervisor
Intel(R)

💣 PoC CODE AVAILABLE

Exploit code for this CVE is tracked in our database. Access poc_code via API Pro.

[ API_PRO ] →
◈ SOURCE Analysis & enrichment: valtersit.com/cve/CVE-2025-35973/ · CVE data: NVD/MITRE (public domain) · © 2026 ValtersIT Hub
>_ SYSTEM_CORE: INTEL_ACTIVE
90% TEAM-CURATED // 10% AUTOMATED INGESTION.
ALL PATCH REFERENCES AND VERSION DATA SOURCED FROM NVD + OSV OFFICIAL REGISTRIES.