VERCEL revealed on 23 April 2026 that it has identified an additional set of customer accounts compromised as part of its security incident, expanding its investigation to include extra compromise indicators and reviews of network requests and environment variable read events.
The company said in an update that a small number of customer accounts showed evidence of prior compromise independent of and predating the incident, possibly resulting from social engineering, malware, or other methods, and that affected parties have been notified though the exact number of customers impacted was not disclosed.
The breach originated with Context[.]ai after it was used by a Vercel employee, enabling the attacker to seize control of the employee’s Google Workspace account and then access the Vercel account, from which they pivoted into the Vercel environment and enumerated and decrypted non-sensitive environment variables.
Hudson Rock later reported that a Context[.]ai employee was infected with Lumma Stealer in February 2026, indicating this may have been the “patient zero” that triggered the malicious actions, with the threat actor active beyond Context[.]ai’s compromise. Vercel CEO Guillermo Rauch said threat intel points to malware distribution to computers in search of tokens like keys to Vercel accounts and other providers.