Nvidia-Backed AI Startup Fireworks AI Hit by Second Cyberattack in Weeks

TEHRAN, Nov 11 (OPA) — Fireworks AI notified customers that unauthorized actors breached the infrastructure of an external cloud service provider on October 24, gaining access to the startup's processing servers. The security breach resulted in the theft of tens of thousands of files, some containing sensitive customer information.

This incident marks the second major security failure for the artificial intelligence startup in a matter of weeks. Earlier in October, the company disclosed a previous intrusion where attackers utilized compromised internal credentials to access user accounts, financial data, and leak sensitive datasets and incomplete models to a limited number of users.

Fireworks AI provides model-serving infrastructure to more than 10,000 enterprise customers, including legal tech firm Harvey. Following the unauthorized access, the startup engaged cybersecurity firm Palo Alto Networks to conduct a forensic investigation and reported the breaches to law enforcement agencies.

The company achieved a $17.5 billion valuation in July and is currently in discussions to raise new capital at a valuation between $30 billion and $40 billion. Semiconductor giants Nvidia and AMD are among the financial backers and supporters of the startup.