Google rolls out improved SynthID AI content detector, now available globally
Summary
Google has launched a new website where anyone can check if images, videos, or audio contain an invisible AI watermark called SynthID. This tool now detects watermarks not only from Google’s AI but also from partners like OpenAI, Nvidia, and Apple, making it easier to identify AI-generated or edited media worldwide.Key Facts
- SynthID is an invisible watermark placed inside AI-created images, videos, and audio by Google’s Gemini AI and its partners.
- Over 180 billion images/videos and 240,000 years of audio have been marked with SynthID so far.
- The new SynthID detector website allows public users to check files for AI watermarks from multiple companies, not just Google.
- Users must log in with Google, OpenAI, or Apple accounts and have a daily limit of about 10 checks to prevent misuse.
- Previously, only Google’s internal testers could detect these watermarks, and the tool only recognized Google’s versions.
- The detector gives a simple yes/no answer on the presence of SynthID but no detailed map of where the watermark appears.
- Some AI content from other sources or open models may not have SynthID or any watermark, so not all AI creations can be identified.
- Google warns that SynthID might be removed or bypassed in the future, so the system is not foolproof.
Read the Full Article
This is a fact-based summary from The Actual News. Click below to read the complete story directly from the original source.
