SynthID – Watermarking and identifying AI-generated text

Channel:
United States Google for Developers
Subscribers:
2,520,000
Published on ● Video Link: https://www.youtube.com/watch?v=_fMFb2Lv7rI


Duration: 0:00
127 views
11

SynthID watermarks and identifies AI-generated content by embedding digital watermarks directly into AI-generated images, audio, text or video. In this talk we will explore SynthID for text, how it works, its open source implementation in HuggingFace Transformers and how to configure and deploy it. Learn how SynthID helps mitigate risks associated with AI-generated content.

Subscribe to Google for Developers → https://goo.gle/developers

#Gemma #GemmaDeveloperDay

Speaker: Tatiana Matejovicova
Products Mentioned: Gemma



Other Videos By Google for Developers


2025-04-06Android Studio tools 🤝 Efficient debugging
2025-04-03Running Gemma on Mac and Windows PCs with Ollama
2025-04-03Let’s rewind it back to Google IO ‘24
2025-04-02Can you beat our time solving the green world? #GoogleIO
2025-04-02Pushing the capabilities of Gemma 3 via distillation and RL fine-tuning
2025-04-02Welcome to the Gemmaverse
2025-04-02Gemma on mobile and web. Best and worst practices
2025-04-02ShieldGemma 2 – Developing safe and responsible AI for images
2025-04-02Integrating the Gemma model family in Transformers
2025-04-02RLHF for finer alignment with Gemma 3
2025-04-02SynthID – Watermarking and identifying AI-generated text
2025-04-02A truly multilingual Gemma 3
2025-04-02PaliGemma – Making Gemma 2 see by adding a vision encoder
2025-04-02Gemma 3 architecture and design
2025-04-02LeRobot – Lowering the entry barrier to AI for robotics
2025-04-01Panel: The importance of open models
2025-04-01Building intelligent agents with Gemma 3
2025-04-01Leveraging large-scale models for efficient learning
2025-04-01Live demo: Gemma 3 on Jetson Orin Nano
2025-04-01Using and fine-tuning Gemma 3
2025-04-01Deepdive into Gemma 3