Vercel Reports Internal Access After Context.ai Compromise

Summary

Vercel says attackers used a compromised Context.ai integration to access an employee’s Google Workspace account and internal systems, exposing some non-sensitive environment variables. The scope and possible exfiltration remain under investigation.

Key points

Vercel says attackers used a compromised Context.ai integration to access an employee’s Google Workspace account and move into Vercel’s internal systems.
The attackers accessed environment variables not marked as sensitive; Vercel says there is no evidence that encrypted sensitive variables were read.
Vercel is still assessing whether data was exfiltrated. It says the number of affected customers appears limited and core services remain operational.
Vercel notified affected customers and advised reviewing account activity, rotating credentials, environment variables and deployment tokens, and auditing deployments.
Vercel published indicators tied to a compromised Google Workspace OAuth application, which it says may have affected other organizations.
Vercel is investigating with Mandiant, Context.ai and law enforcement, and says it has added protections and improved dashboard controls for environment variables.

Article Details

Victim Organization: Vercel
Incident Type: Unauthorized access to Vercel internal systems following the compromise of a third-party AI tool and access to an employee’s Google Workspace account.
Data Types Exposed: Customer credentials (a subset was found to be compromised; the specific credentials and number of affected customers were not disclosed)
Environment variables not marked as sensitive (accessed; Vercel reported no evidence that encrypted sensitive environment variables were read)
Affected Records: A limited, unspecified number of customers; Vercel said the impact appeared limited.
Operational Impact: Attackers gained unauthorized access to certain internal systems and accessed some environment variables. Core services remained operational. Vercel said it was still assessing whether data had been exfiltrated.
Ransom Or Extortion: Unknown
Claim Status: confirmed

People

Guillermo RauchVercel CEO who shared information about the attacker’s steps to escalate access from the compromised account into Vercel environments.

Vendors

Context.aiVercel disclosed an incident where attackers gained unauthorized access to internal systems after a third-party AI tool, Context.ai, was compromised and used to access an employee’s Google Workspace account.GoogleVercel disclosed an incident where attackers gained unauthorized access to internal systems after a third-party AI tool, Context.ai, was compromised and used to access an employee’s Google Workspace account.MandiantThe company says the impact appears limited, has notified affected users and advised credential and environment-variable rotation while working with Mandiant and law enforcement and publishing indicators of compromise.VercelVercel disclosed an incident where attackers gained unauthorized access to internal systems after a third-party AI tool, Context.ai, was compromised and used to access an employee’s Google Workspace account.

Products

Context.aiVercel disclosed an incident where attackers gained unauthorized access to internal systems after a third-party AI tool, Context.ai, was compromised and used to access an employee’s Google Workspace account.Google WorkspaceVercel disclosed an incident where attackers gained unauthorized access to internal systems after a third-party AI tool, Context.ai, was compromised and used to access an employee’s Google Workspace account.VercelVercel disclosed an incident where attackers gained unauthorized access to internal systems after a third-party AI tool, Context.ai, was compromised and used to access an employee’s Google Workspace account.