GCP Vertex AI Caller Impossible Travel

Last updated a month ago on 2026-10-07
Created 2 months ago on 2026-10-02

About

Detects the same Vertex AI authenticated caller (client.user.email) generating aiplatform audit activity from geographically distant source IPs within a short window at an impossible travel speed. That pattern can indicate a stolen user or service-account credential used from multiple regions. Related: GCP Vertex AI Caller Activity From High Number of Countries covers slower multi-country reuse without a high-speed hop.
Tags
Domain: GenAIDomain: CloudData Source: GCP Vertex AIData Source: GCPData Source: Google Cloud PlatformPlatform: GCPService: GCP Vertex AIUse Case: Threat DetectionTactic: Credential AccessTactic: Initial AccessThreat: Impossible TravelThreat: LLMjackingThreat: Unauthorized AI UsageMitre Atlas: AML.T0012Mitre Atlas: AML.T0091Mitre Atlas: AML.T0091.000Rule Type: ES|QLLanguage: esql
Severity
medium
Risk Score
47
MITRE ATT&CK™

Credential Access (TA0006)(external, opens in a new tab or window)

Initial Access (TA0001)(external, opens in a new tab or window)

False Positive Examples
Distributed CI or multi-region automation that shares one service account across distant egress IPs. Prefer per-region identities, or raise distance and speed thresholds. VPN or cloud egress that geolocates incorrectly or jumps between distant POPs. Validate the IP locations before rotating credentials.
License
Elastic License v2(external, opens in a new tab or window)

Definition

Integration Pack
Prebuilt Security Detection Rules
Related Integrations

gcp_vertexai(external, opens in a new tab or window)

Query
text code block:
from logs-gcp_vertexai.auditlogs-* | where data_stream.dataset == "gcp_vertexai.auditlogs" and client.user.email is not null and source.ip is not null and source.geo.location is not null and source.geo.country_name is not null and ( event.action like "*PredictionService.GenerateContent*" or event.action like "*PredictionService.StreamGenerateContent*" or event.action like "*PredictionService.CountTokens*" ) | eval Esql.source_geo_lat = st_y(source.geo.location), Esql.source_geo_lon = st_x(source.geo.location) | where Esql.source_geo_lat is not null and Esql.source_geo_lon is not null | stats Esql.first_lat = first(Esql.source_geo_lat, @timestamp), Esql.first_lon = first(Esql.source_geo_lon, @timestamp), Esql.last_lat = last(Esql.source_geo_lat, @timestamp), Esql.last_lon = last(Esql.source_geo_lon, @timestamp), Esql.event_count = count(*), Esql.country_count = count_distinct(source.geo.country_name), Esql.source_ip_values = values(source.ip), Esql.source_geo_country_name_values = values(source.geo.country_name), Esql.source_geo_region_name_values = values(source.geo.region_name), Esql.source_geo_city_name_values = values(source.geo.city_name), Esql.source_as_organization_name_values = values(source.as.organization.name), Esql.event_action_values = values(event.action), Esql.user_agent_original_values = values(user_agent.original), Esql.timestamp_first_seen = min(@timestamp), Esql.timestamp_last_seen = max(@timestamp) by client.user.email | where Esql.event_count >= 2 and Esql.country_count >= 2 | eval Esql.p1 = to_geopoint(concat("POINT(", to_string(Esql.first_lon), " ", to_string(Esql.first_lat), ")")), Esql.p2 = to_geopoint(concat("POINT(", to_string(Esql.last_lon), " ", to_string(Esql.last_lat), ")")) | eval Esql.distance_km = round(st_distance(Esql.p1, Esql.p2) / 1000.0, 0), Esql.window_minutes = date_diff("minute", Esql.timestamp_first_seen, Esql.timestamp_last_seen), Esql.travel_kmh = case(Esql.window_minutes > 0, round(Esql.distance_km * 60.0 / Esql.window_minutes, 0), null), source.ip = MV_FIRST(Esql.source_ip_values) | where Esql.distance_km >= 500 and Esql.travel_kmh >= 800 | keep client.user.email, source.ip, Esql.source_ip_values, Esql.source_geo_country_name_values, Esql.source_geo_region_name_values, Esql.source_geo_city_name_values, Esql.source_as_organization_name_values, Esql.event_action_values, Esql.user_agent_original_values, Esql.distance_km, Esql.travel_kmh, Esql.window_minutes, Esql.country_count, Esql.event_count, Esql.timestamp_first_seen, Esql.timestamp_last_seen

Install detection rules in Elastic Security

Detect GCP Vertex AI Caller Impossible Travel in the Elastic Security detection engine by installing this rule into your Elastic Stack.

To setup this rule, check out the installation guide for Prebuilt Security Detection Rules(external, opens in a new tab or window).