From deepfakes to DNA: the science of watermarking AI // TRAIN BRAIN

From deepfakes to DNA: the science of watermarking AI

How do you know if what you're seeing is real? Professor Hannah Fry sits down with Pushmeet Kohli (VP of Science) and Jeremy Ratcliffe (Research Scientist) to break down SynthID, the watermarking technology expanding from text, images, and video straight into biology. Learn how invisible, mathematical watermarks help trace AI provenance, protect against misuse, and safeguard protein sequence design without impacting biological function.
Timecodes:
00:00 Introduction
00:34 What is a watermark?
04:35 SynthID
15:16 Images and video
19:28 SynthID Bio
28:00 Future of resilience
Learn more about SynthID Bio: https://deepmind.google/blog/introducing-synthid-bio/
___
Subscribe to our channel https://www.youtube.com/@googledeepmind
Find us on X https://x.com/GoogleDeepMind
Follow us on Instagram https://instagram.com/googledeepmind
Add us on Linkedin https://www.linkedin.com/company/deepmind/

Google DeepMind

Artificial intelligence could be one of humanity's most useful inventions. DeepMind aims to build advanced AI to expand our knowledge and find new answers. By solving this one thing, we believe we could help people solve thousands of problems. We’re a te...

From deepfakes to DNA: the science of watermarking AI

Can AI help us better predict the weather?

AlphaGenome Atlas: Using AI to find disease causing variants ?

AlphaGenome Atlas: Understanding the human genome

WeatherNext 3: More accurate, timely, and local weather forecasts

The mathematics of AI uncertainty

Talking to Apollo about Gemini Robotics 2 ?

This is Gemini Robotics 2 ?

Robots working together with Gemini Robotics 2

Intelligent whole-body control with Gemini Robotics 2

Multi-robot collaboration with Gemini Robotics 2

Advanced dexterity with Gemini Robotics 2

Gemini Robotics 2 brings whole body intelligence to robots

Understanding the inner thoughts of AI

When millions of AI agents meet

Gemini for Science is here. ?

SynthID, our imperceptible watermark for AI-generated content, is expanding to more partners.

Gemini 3.5 Flash has landed.

Build your next story with Gemini Omni.

Generating novel scientific hypotheses with Co-Scientist

Using AI to outsmart drug-resistant bacteria

Understanding cancer at a genetic level with AI

Predicting a historic storm earlier with WeatherNext

Reimagining a 50-year-old interface (the mouse pointer) with AI

What’s new in Gemma 4?

Teaching the foundations of AI in the classroom

“Dear Upstairs Neighbors” (BTS)

Gemini 3 Flash: Transform messy data

Gemini 3 Flash: Evolve code faster

Gemini 3 Flash: Orchestrate a function call kitchen

Gemini 3 Flash: Generate a narrative of your journey

The future of intelligence | Demis Hassabis (Co-founder and CEO of DeepMind)

The arrival of AGI | Shane Legg (co-founder of DeepMind)

Robotics lab tour with Hannah Fry | Bonus episode!

What does ancient bridge building and AI have in common?

What’s new in Gemini 3?

AlphaFold: Grand challenge to Nobel Prize with John Jumper

The Thinking Game | Available to watch for free on @googledeepmind

AlphaFold: The 50-year grand challenge cracked by AI

The Thinking Game | Full documentary | Tribeca Film Festival official selection

The Thinking Game | Documentary trailer

Nano Banana Pro: Your new creative partner

Vibe Coding with Gemini 3 in Google AI Studio

Gemini 3: Reasoning with voxel art

A new era of intelligence with Gemini 3

WeatherNext 2: Our most advanced weather forecasting model

SIMA 2: An agent that plays, reasons, and learns with you in virtual 3D worlds

Part 2: Social engineering, malware, and the future of cybersecurity in AI

Veo 3.1 - Add and remove objects to your scene

Veo 3.1 - Frames to video

Veo 3.1 - Ingredients to video

Veo 3.1 - Create longer, seamless shots

Veo 3.1 - Designed to empower creatives

Beyond phishing: Cyber threats in the age of AI with Four Flynn (pt. 1)

From sketches to prototype: Designing with generative AI

Gemini Robotics 1.5: Using agentic capabilities

Gemini Robotics 1.5: Learning across embodiments

Gemini Robotics 1.5: Enabling robots to plan, think and use tools to solve complex tasks

Gemini Robotics 1.5: Thinking while acting

Google DeepMind researchers react to Nano Banana demos ?

Genie 3: An infinite world model with Shlomi Fruchter and Jack Parker-Holder

Can AI help to save endangered birds?

Genie 3: Creating dynamic worlds that you can navigate in real life

Solving years-old math problems with Gemini 2.5 Deep Think

A whistle stop tour of AI creation with Paige Bailey

AI and the Future of Health with Joelle Barral

Should you say ‘please’ when using AI?

Consciousness, Reasoning and the Philosophy of AI with Murray Shanahan

Using AI to create music ?

Can AI go beyond human knowledge?

Is Human Data Enough? With David Silver

Thomas' story

Creating with Veo 2

Gemini Robotics: Generality in action

Project Astra: Exploring a Universal AI Assistant with Greg Wayne

Gemini 2.0 and the evolution of agentic AI with Oriol Vinyals

Project Astra early access demo | Using AI as a creative tool

Project Astra early access demo | Using AI for better marathon training

Gemini 2.0 for games demo | Playing Squad Busters, Clash of Clans, and Hay Day

Gemini 2.0 for games demo | Playing Hay Day