Posted 4/12/2024, 3:03:16 PM
Google Deploys New AI to Block More Gmail Spam, Detect Confidential Data
- Google has deployed custom large language models (LLMs) to improve Gmail spam and phishing detection, blocking 20% more spam
- The LLMs review 1000% more user-reported Gmail spam per day and respond 90% faster to new spam/phishing attacks in Drive
- The LLMs stop 99.9% of spam and detect twice as much malware as third-party products
- Google is focused on using AI to tackle the remaining 0.1% of spam that gets through
- Google offers an optional new AI tool for $10 per user/month to automatically find and protect confidential data in files