CVE-2026-48166: filament timing side-channel

CISO Take

Filament's login page contains an observable timing discrepancy (CWE-208) that lets any unauthenticated attacker over the network determine whether a given email address has a registered account. While the direct impact is bounded to information disclosure (CVSS 5.3 Medium, Confidentiality:Low), the risk escalates when Filament serves as the admin interface for AI/ML platforms — enumerated administrator and ML engineer emails become targeted inputs for credential stuffing and spearphishing campaigns against personnel with privileged access to model deployments and training infrastructure. No public exploits exist, EPSS data is unavailable, and this CVE is absent from the CISA KEV, indicating no confirmed active exploitation at this time. Upgrade to Filament 4.11.5 or 5.6.5 immediately; if patching is delayed, add consistent-time response middleware at the proxy layer to eliminate the observable timing delta.

Sources: NVD GitHub Advisory ATLAS

What is the risk?

Medium risk in general web contexts, elevated when Filament administers AI/ML platforms. The vulnerability requires no authentication, no user interaction, and is exploitable from the network with low complexity, making automated email enumeration mechanically trivial. Direct impact is bounded to information disclosure — no code execution, training data poisoning, or model tampering is reachable through this vector alone. However, enumerated administrator or ML engineer email addresses are high-value reconnaissance assets for follow-on attacks targeting AI system access. The absence of public exploits and KEV listing reduces immediate urgency, but the simplicity of timing-based HTTP enumeration means any attacker capable of scripting HTTP requests can weaponize it without specialized tooling.

How does the attack unfold?

Active Scanning

Attacker scripts automated HTTP POST requests to the Filament login endpoint, cycling through a wordlist of target email addresses and recording precise response times for each.

AML.T0006

Identity Enumeration

Timing delta between registered accounts (bcrypt comparison performed, ~150-200ms) and unregistered addresses (early return, ~30-50ms) reveals which emails have active accounts.

AML.T0087

Target List Construction

Attacker compiles a verified list of confirmed administrator and ML engineer email addresses associated with the AI platform, filtering false positives via repeated measurements.

AML.T0087

Credential or Social Attack

Enumerated emails are weaponized for credential stuffing against the AI admin panel or fed into spearphishing campaigns targeting personnel with model deployment and API key access.

AML.T0012

Active Scanning

Attacker scripts automated HTTP POST requests to the Filament login endpoint, cycling through a wordlist of target email addresses and recording precise response times for each.

AML.T0006

Identity Enumeration

Timing delta between registered accounts (bcrypt comparison performed, ~150-200ms) and unregistered addresses (early return, ~30-50ms) reveals which emails have active accounts.

AML.T0087

Target List Construction

Attacker compiles a verified list of confirmed administrator and ML engineer email addresses associated with the AI platform, filtering false positives via repeated measurements.

AML.T0087

Credential or Social Attack

Enumerated emails are weaponized for credential stuffing against the AI admin panel or fed into spearphishing campaigns targeting personnel with model deployment and API key access.

AML.T0012

What systems are affected?

Package	Ecosystem	Vulnerable Range	Patched
filament	—	—	No patch

Do you use filament? You're affected.

How severe is it?

CVSS 3.1

5.3 / 10

EPSS

N/A

Exploitation Status

No known exploitation

Sophistication

Trivial

What is the attack surface?

AV Network

AC Low

PR None

UI None

S Unchanged

C Low

I None

A None

What should I do?

1 step

1) Patch immediately: upgrade Filament to 4.11.5 (4.x branch) or 5.6.5 (5.x branch) — the fix implements constant-time login response handling to eliminate the timing signal. 2) If patching is deferred, configure a reverse proxy or WAF rule to pad all login endpoint responses to a fixed minimum duration (e.g., 500ms), neutralizing observable timing deltas before they reach the attacker. 3) Enforce aggressive rate limiting on the login endpoint (e.g., 5 attempts per minute per IP or per email) to slow automated enumeration campaigns. 4) Deploy CAPTCHA or login attempt throttling as a compensating control on high-risk AI platform admin interfaces. 5) Monitor login endpoint access logs for repeated requests with systematically varied email addresses from the same source IP — a characteristic signature of automated timing enumeration tools.

How is it classified?

Privacy Violation Auth Bypass Framework AML.T0006 - Active Scanning AML.T0012 - Valid Accounts AML.T0087 - Gather Victim Identity Information

Which compliance frameworks are affected?

This CVE is relevant to:

EU AI Act

Article 9 - Risk management system

ISO 42001

A.9.4 - Access to systems and applications

NIST AI RMF

GOVERN 1.7 - Processes for AI risk management

OWASP LLM Top 10

LLM06 - Sensitive Information Disclosure

Frequently Asked Questions

What is CVE-2026-48166?

Filament's login page contains an observable timing discrepancy (CWE-208) that lets any unauthenticated attacker over the network determine whether a given email address has a registered account. While the direct impact is bounded to information disclosure (CVSS 5.3 Medium, Confidentiality:Low), the risk escalates when Filament serves as the admin interface for AI/ML platforms — enumerated administrator and ML engineer emails become targeted inputs for credential stuffing and spearphishing campaigns against personnel with privileged access to model deployments and training infrastructure. No public exploits exist, EPSS data is unavailable, and this CVE is absent from the CISA KEV, indicating no confirmed active exploitation at this time. Upgrade to Filament 4.11.5 or 5.6.5 immediately; if patching is delayed, add consistent-time response middleware at the proxy layer to eliminate the observable timing delta.

Is CVE-2026-48166 actively exploited?

No confirmed active exploitation of CVE-2026-48166 has been reported, but organizations should still patch proactively.

How to fix CVE-2026-48166?

1) Patch immediately: upgrade Filament to 4.11.5 (4.x branch) or 5.6.5 (5.x branch) — the fix implements constant-time login response handling to eliminate the timing signal. 2) If patching is deferred, configure a reverse proxy or WAF rule to pad all login endpoint responses to a fixed minimum duration (e.g., 500ms), neutralizing observable timing deltas before they reach the attacker. 3) Enforce aggressive rate limiting on the login endpoint (e.g., 5 attempts per minute per IP or per email) to slow automated enumeration campaigns. 4) Deploy CAPTCHA or login attempt throttling as a compensating control on high-risk AI platform admin interfaces. 5) Monitor login endpoint access logs for repeated requests with systematically varied email addresses from the same source IP — a characteristic signature of automated timing enumeration tools.

What systems are affected by CVE-2026-48166?

This vulnerability affects the following AI/ML architecture patterns: ML platform admin dashboards, Laravel-based AI SaaS applications, Web application backends.

What is the CVSS score for CVE-2026-48166?

CVE-2026-48166 has a CVSS v3.1 base score of 5.3 (MEDIUM).

What is the AI security impact?

Affected AI Architectures

ML platform admin dashboardsLaravel-based AI SaaS applicationsWeb application backends

MITRE ATLAS Techniques

AML.T0006 Active Scanning

AML.T0012 Valid Accounts

AML.T0087 Gather Victim Identity Information

Compliance Controls Affected

EU AI Act: Article 9

ISO 42001: A.9.4

NIST AI RMF: GOVERN 1.7

OWASP LLM Top 10: LLM06

What are the technical details?

Original Advisory

Filament is a collection of full-stack components for accelerated Laravel development. From 4.0.0 until 4.11.5 and 5.6.5, the login page has an observable timing discrepancy that allows unauthenticated attackers to enumerate registered email addresses. The impact is limited to disclosing whether an account exists for a given email. This vulnerability is fixed in 4.11.5 and 5.6.5.

Exploitation Scenario

An adversary targeting a Laravel-based AI SaaS platform writes a simple script that sends thousands of HTTP POST requests to the Filament login endpoint, each with a different email address drawn from a wordlist of corporate email patterns, LinkedIn-scraped names, or breach datasets. By measuring response latency, the attacker distinguishes registered accounts — where the backend performs a bcrypt hash comparison, adding ~150-200ms — from unregistered accounts that return immediately after a failed lookup. Within minutes, the attacker has a verified list of administrator and ML engineer email addresses. These are cross-referenced against public breach databases for plaintext passwords and submitted to the AI platform's login page in a credential stuffing campaign, or used to craft highly convincing spearphishing emails impersonating IT or security teams, targeting individuals whose credentials would unlock model management consoles, training job controls, and API key generation.

Weaknesses (CWE)

CWE-208 Observable Timing Discrepancy Primary CWE-208 Observable Timing Discrepancy

CWE-208 — Observable Timing Discrepancy: Two separate operations in a product require different amounts of time to complete, in a way that is observable to an actor and reveals security-relevant information about the state of the product, such as whether a particular operation was successful or not.

Source: MITRE CWE corpus.