OpenAI ChatGPT Enterprise Multiple Authentication Failures
Description
Detects multiple failed ChatGPT Enterprise login attempts for the same user email within a short window. Password spraying or credential stuffing against enterprise accounts often precedes a successful compromise.
Query · esql
from logs-openai_chatgpt_enterprise.auth_log-*
| where
data_stream.dataset == "openai_chatgpt_enterprise.auth_log" and
event.action == "login_failed" and
user.email is not null
| stats
Esql.event_count = count(*),
Esql.source_ip_values = values(source.ip),
Esql.source_geo_country_iso_code_values = values(source.geo.country_iso_code),
Esql.user_agent_original_values = values(user_agent.original),
Esql.timestamp_first_seen = min(@timestamp),
Esql.timestamp_last_seen = max(@timestamp)
by user.email
| where Esql.event_count >= 10
| keep user.email, Esql.*
Investigation fields
Pivot points the source recommends for triage.
user.emailEsql.event_countEsql.source_ip_valuesEsql.source_geo_country_iso_code_valuesEsql.timestamp_first_seenEsql.timestamp_last_seen
Implementation guide
This rule requires the OpenAI ChatGPT Enterprise integration to collect compliance logs into Elasticsearch.
https://www.elastic.co/docs/reference/integrations/openai_chatgpt_enterprise
Known false positives
- Users mistype passwords repeatedly or SSO misconfiguration can generate clusters of failed logins.
Analyst notes
Investigating OpenAI ChatGPT Enterprise Multiple Authentication Failures
One email produced a burst of failed ChatGPT Enterprise login attempts. That pattern can indicate password spraying, credential stuffing, or repeated guessing before a successful compromise.
Escalate when many distinct source IPs are involved, failures precede a successful login from an unfamiliar IP or user agent, or privileged activity follows. Close as a false positive for repeated typos or SSO misconfiguration from a single client.
Possible investigation steps
- Review
Esql.event_count,Esql.source_ip_values, and geo values on the alert. - Low IP diversity usually indicates one device or egress. High diversity suggests distributed automation.
- Search for successful logins for the same email immediately after the failure burst.
- Correlate with role changes, invites, or data-access events if a successful login follows.
False positive analysis
- Users mistyping passwords or temporary SSO misconfiguration can generate short clusters of failures from one IP and user agent.
Response and remediation
- If compromise is suspected, lock or reset the account, revoke sessions, and review subsequent successful authentication and audit activity.