Detection rules › Splunk

AWS Bedrock Delete Model Invocation Logging Configuration

Status
production
Severity
medium
Group by
aws::awsRegion, aws::recipientAccountId, aws::userAgent, dest, signature, src, user, vendor_product
Author
Bhavin Patel, Splunk
Source
github.com/splunk/security_content

The following analytic identifies attempts to delete AWS Bedrock model invocation logging configurations. It leverages AWS CloudTrail logs to detect when a user or service calls the DeleteModelInvocationLogging API. This activity is significant as it may indicate an adversary attempting to remove audit trails of model interactions after compromising credentials. Deleting model invocation logs could allow attackers to interact with AI models without leaving traces, potentially enabling them to conduct data exfiltration, prompt injection attacks, or other malicious activities without detection. If confirmed malicious, this could represent a deliberate attempt to hide unauthorized model usage and evade detection.

Known false positives

  • Legitimate administrators may delete model invocation logging configurations during maintenance, when updating logging policies, or when cleaning up unused resources. Consider implementing an allowlist for expected administrators who regularly manage logging configurations.

MITRE ATT&CK coverage

Telemetry coverage

Rules detecting the same action

These rules filter on the same operation.

Rule body

name: AWS Bedrock Delete Model Invocation Logging Configuration
id: 9c5e3d62-f743-11ee-9f6e-acde48001124
version: 7
creation_date: '2025-03-25'
modification_date: '2026-05-13'
author: Bhavin Patel, Splunk
status: production
type: TTP
description: The following analytic identifies attempts to delete AWS Bedrock model invocation logging configurations. It leverages AWS CloudTrail logs to detect when a user or service calls the DeleteModelInvocationLogging API. This activity is significant as it may indicate an adversary attempting to remove audit trails of model interactions after compromising credentials. Deleting model invocation logs could allow attackers to interact with AI models without leaving traces, potentially enabling them to conduct data exfiltration, prompt injection attacks, or other malicious activities without detection. If confirmed malicious, this could represent a deliberate attempt to hide unauthorized model usage and evade detection.
data_source:
    - AWS CloudTrail DeleteModelInvocationLoggingConfiguration
search: >-
    `cloudtrail` eventSource=bedrock.amazonaws.com
    eventName=DeleteModelInvocationLoggingConfiguration | rename user_name as user
    | stats count min(_time) as firstTime max(_time) as lastTime by src user
    user_agent vendor_account vendor_product dest signature vendor_region |
    `security_content_ctime(firstTime)` | `security_content_ctime(lastTime)` |
    `aws_bedrock_delete_model_invocation_logging_configuration_filter`
how_to_implement: The Splunk AWS Add-on is required to utilize this data. The search requires AWS CloudTrail logs with Bedrock service events enabled. You must install and configure the AWS App for Splunk (version 6.0.0 or later) and Splunk Add-on for AWS (version 5.1.0 or later) to collect CloudTrail logs from AWS. Ensure the CloudTrail is capturing Bedrock model invocation logging management events.
known_false_positives: Legitimate administrators may delete model invocation logging configurations during maintenance, when updating logging policies, or when cleaning up unused resources. Consider implementing an allowlist for expected administrators who regularly manage logging configurations.
references:
    - https://www.sumologic.com/blog/defenders-guide-to-aws-bedrock/
    - https://attack.mitre.org/techniques/T1562/008/
finding:
    title: User $user$ deleted AWS Bedrock model invocation logging from $src$
    entity:
        field: user
        type: user
        score: 50
threat_objects:
    - field: src
      type: ip_address
analytic_story:
    - AWS Bedrock Security
asset_type: AWS Account
mitre_attack_id:
    - T1685.002
product:
    - Splunk Enterprise
    - Splunk Enterprise Security
    - Splunk Cloud
category: cloud
security_domain: threat

Stages and Predicates

Stage 1: search

`cloudtrail` eventSource=bedrock.amazonaws.com eventName=DeleteModelInvocationLoggingConfiguration

Stage 2: rename

| rename user_name as user

Stage 3: stats

| stats count min(_time) as firstTime max(_time) as lastTime by src user user_agent vendor_account vendor_product dest signature vendor_region

Stage 4: search

| `security_content_ctime(firstTime)`

Stage 5: search

| `security_content_ctime(lastTime)`

Stage 6: search

| `aws_bedrock_delete_model_invocation_logging_configuration_filter`

Indicators

These rows show field, operator, and value matches.