Fireworks AI gets burned twice as customer data exposure widens
Inference biz pins latest breach on cloud provider and LLMJacking miscreants
Fireworks AI announced a cyberattack in October that led to wider exposure of customer data, including billing records, datasets, and incomplete model weights. The breach occurred when attackers compromised a third-party cloud provider on October 23, granting them access to Fireworks' development systems the next day. Suspecting financial motivations, Fireworks attributed the incident to LLMJacking, the selling of unauthorized AI model access.
The compromised data fell into three categories: resource metadata, billing information, and model weights/datasets. While payment card numbers remained unaffected, some customers experienced partial and incomplete downloads of their data. Fireworks co-founder and CEO Lin Qiao emphasized that the attackers were not after model weights, addressing concerns from Cantina Security CEO Hari Mulackal.
Render CEO Anurag Goel reported that some chats with their AI assistant may have been exposed in the incident. Notion's security team confirmed the Render workspace as part of the compromised systems.
Written by urgent.news from The Register Science's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
This story
This is one outlet's version. Read the fullest account.
