[New Rule] Potential Abuse of Resources by High Token Count and Large Response Sizes (#3644)

detection_rules/schemas/definitions.py

@@ -88,6 +88,7 @@ EXPECTED_RULE_TAGS = [
     'Domain: Cloud',
     'Domain: Container',
     'Domain: Endpoint',
+    'Mitre Atlas: *',
     'OS: Linux',
     'OS: macOS',
     'OS: Windows',

rules/integrations/aws/aws_bedrock_high_resource_consumption_detection.toml

@@ -0,0 +1,58 @@
+[metadata]
+creation_date = "2024/05/04"
+maturity = "production"
+updated_date = "2024/05/04"
+min_stack_comments = "ES|QL rule type is still in technical preview as of 8.13, however this rule was tested successfully; integration in tech preview"
+min_stack_version = "8.13.0"
+
+[rule]
+author = ["Elastic"]
+description = """
+Detects potential resource exhaustion or data breach attempts by monitoring for users who consistently generate high input token counts, submit numerous requests, and receive
+large responses. This behavior could indicate an attempt to overload the system or extract an unusually large amount of data, possibly revealing sensitive information or
+causing service disruptions.
+"""
+false_positives = ["Authorized heavy usage of the system that is business justified and monitored."]
+from = "now-60m"
+interval = "10m"
+language = "esql"
+license = "Elastic License v2"
+name = "Potential Abuse of Resources by High Token Count and Large Response Sizes"
+references = [
+    "https://atlas.mitre.org/techniques/AML.T0051",
+    "https://owasp.org/www-project-top-10-for-large-language-model-applications/",
+    "https://www.elastic.co/security-labs/elastic-advances-llm-security",
+]
+risk_score = 47
+rule_id = "b1773d05-f349-45fb-9850-287b8f92f02d"
+setup = """## Setup
+
+This rule requires that guardrails are configured in AWS Bedrock. For more information, see the AWS Bedrock documentation:
+
+https://docs.aws.amazon.com/bedrock/latest/userguide/guardrails-create.html
+"""
+severity = "medium"
+tags = [
+    "Domain: LLM",
+    "Data Source: AWS Bedrock",
+    "Data Source: Amazon Web Services",
+    "Data Source: AWS S3",
+    "Use Case: Potential Overload",
+    "Use Case: Resource Exhaustion",
+    "Mitre Atlas: LLM04"
+]
+timestamp_override = "event.ingested"
+type = "esql"
+
+query = '''
+from logs-aws_bedrock.invocation-*
+| stats max_tokens = max(gen_ai.usage.prompt_tokens),
+         total_requests = count(*),
+         avg_response_size = avg(gen_ai.usage.completion_tokens)
+  by user.id
+// tokens count depends on specific LLM, as is related to how embeddings are generated.
+| where max_tokens > 5000 and total_requests > 10 and avg_response_size > 500
+| eval risk_factor = (max_tokens / 1000) * total_requests * (avg_response_size / 500)
+| where risk_factor > 10
+| sort risk_factor desc
+'''