- Horizon AI
- Posts
- Denmark’s Surprising Approach to AI Deepfakes 👀
Denmark’s Surprising Approach to AI Deepfakes 👀
Get Claude to write like you do
Gina Acosta
June 27, 2025
Welcome to another edition of Horizon AI,
As AI technology gets more advanced, the deepfake problem keeps getting worse. Many policies are in the works to address it, but Denmark has taken a unique approach to tackle the issue.
Let’s get into it!
Read Time: 4.5 min
Here's what's new today in the Horizon AI
Denmark to Tackle Deepfakes by Giving Everyone Copyright Over Their Own Features
Google Launches Gemma 3n, an AI Model Built for Mobile Devices
AI Tutorial: Get Claude to write like you do
AI Tools to check out
AI Findings/Resources
The Latest in AI and Tech 💡
AI News
AI REGULATION
Denmark to Tackle Deepfakes by Giving Everyone Copyright Over Their Own Features
The Danish government is proposing a copyright law amendment to give citizens ownership rights over their body, facial features, and voice.
Details:
This would allow people to demand that companies remove fake images, videos, or audio that mimic them without consent.
Jakob Engel-Schmidt, Denmark’s culture minister, says the change is necessary because current laws are failing to protect “people against generative AI.”
The rule would also cover “realistic, digitally generated imitations” of an artist’s work but would not apply to parodies or satire.
The Danish government proudly refers to it as the first law of its kind in Europe. It comes at a time when rapidly developing AI technology has made it easier than ever to create convincing deepfakes of people without their consent.
Google Launches Gemma 3n, an AI Model Built for Mobile Devices
Google has introduced Gemma 3n, a new family of models designed for real-time use on mobile devices
Details:
It supports images, audio, video and text inputs and text outputs.
Offered in two sizes: E2B (5 B parameters, needs 2 GB RAM) and E4B (8 B parameters, needs 3 GB RAM)
For image and video tasks, Gemma 3n uses a new MobileNet-V5-300M encoder, capable of handling resolutions up to 768×768 pixels and processing up to 60 images per second.
The E4B variant scores over 1300 points in the LMArena benchmark, setting a new record for models under 10 billion parameters.
The model is available for download on Hugging Face and Kaggle, and can also be deployed directly through Google AI Studio, Cloud Run, or Vertex AI.
Google is also offering a Gemma 3n Impact Challenge, looking for applications that use the model’s multimodal and offline capabilities to address real-world problems, with a total prize pool of $150,000.
AI Tutorial
Get Claude to write like you do
Claude allows you to input instructions to fully copy your writing style.
Go to Claude’s website and sign up.
Click ‘Choose Style,’ then select ‘Create and Edit Styles.’
Click ‘Create Custom Style.’
Upload a writing sample in your preferred style so Claude can analyze it properly.
Click ‘Create Style’ and wait for it to process.
Once it’s ready, click ‘Use Style’ to start using it.
Claude will respond in the writing style you provided within seconds.
AI Tools to check out
🌐 Wonderish: Build websites, landing pages, and apps by simply typing. No code, no setup.
🧠 Thunai: An Agentic AI Platform with a self-learning brain that turns your org’s knowledge into smart agents.
🎨 Midjourney: Create images, paintings, digital art, logos and much more simply by writing a prompt.
💥 Daily.co: Real-time voice, video, and AI for developers.
🔍 Perplexity: A free AI-powered answer engine that provides accurate, real-time answers to any question.
AI Findings/Resources
⭐ Top 10 emerging technologies of 2025, according to the World Economic Foru
🤔 AI hallucinates more frequently as it gets more advanced — is there any way to stop it from happening, and should we even try?
⏳ 9 routine tasks ChatGPT can handle in seconds, saving you hours
The latest in AI and Tech
After pausing the launch of Google Photos’ AI-powered 'Ask Photos' tool earlier this month, Google is now resuming the rollout. The feature, which lets users search their photo library using natural language, is now faster and should deliver more relevant results.
For certain searches, YouTube Premium subscribers will now see a curated video carousel at the top. The feature is currently only available on iOS and Android, for English-language videos, and will be open for testing until July 30.
YouTube also announced that its conversational AI tool will be expanding to some non-Premium users in the U.S.
The Berlin Commissioner for Data Protection has reported the AI app to Apple and Google, claiming that it sends user data to China in a way that doesn't comply with European Union law. The two tech firms must now review the report and decide whether to remove the app.
The company would acquire Play AI’s technology and some of its employees if the deal goes through. The startup has raised $23.5 million in total, and its technology can mimic a user's voice or generate human-like voices for AI-powered use cases such as customer service.
That’s a wrap!
We'd love to hear your thoughts on today's email!Your feedback helps us improve our content |
Not subscribed yet? Sign up here and send it to a colleague or friend!
See you in our next edition!
Gina 👩🏻💻