Google has announced that it uses TensorFlow, an open-source machine learning (ML) framework developed by Google, to block 99.9% of spam, phishing, and malware before it reaches Google's 1.5 billion Gmail customers each month. New improvements to protections powered by TensorFlow allow Google to detect and block an additional 100 million spam messages every day. The additional spam messages come from sources such as image-based messages, hidden embedded content, and newly created domains that seek to hide low volume spam campaigns behind legitimate traffic. ML makes catching spam possible by helping us identify patterns in large data sets that humans who create the rules might not catch; it makes it easy for us to adapt quickly to ever-changing spam attempts. By complementing our existing ML models with TensorFlow, we're able to refine these models even further, while allowing the team to focus less on the underlying ML framework, and more on solving the problem: ridding your inbox of spam!