How countries are regulating AI in 2025

PLUS: More tech updates from Meta and Anthropic

In partnership with

AI News Universe

Today’s Insights:

  • How countries are regulating AI in 2025

  • Meta may halt AI projects deemed too risky

  • AI to speed up UK breast cancer screenings

  • Anthropic dares hackers to break its AI

  • Meta’s smart glasses spending nears $100B

There’s a reason 400,000 professionals read this daily.

Join The AI Report, trusted by 400,000+ professionals at Google, Microsoft, and OpenAI. Get daily insights, tools, and strategies to master practical AI skills that drive results.

AI News Today

🌍 How countries are regulating AI in 2025 ↗️LINK

  • What: Countries are adopting different AI regulations. The U.S., under Trump, has scrapped Biden-era oversight, leaving AI largely unregulated. The EU has enacted the world’s most comprehensive AI law, banning high-risk uses like biometric profiling. China controls AI to align with state interests, while India and Britain focus on economic growth over strict laws.

  • Why: Governments must balance innovation with risks like misinformation, bias, and national security threats. The EU prioritizes consumer protection, while China enforces censorship. India and Britain see AI as an economic driver, preferring minimal restrictions.

  • Impact: The global AI divide is widening. The EU's strict laws could set global standards, while the U.S.’s hands-off approach may accelerate AI advancements but increase risks. China’s controlled AI ecosystem ensures government dominance, while India and Britain remain in regulatory limbo, awaiting clearer policies.

⚠️ Meta may halt AI projects deemed too risky ↗️LINK

  • What: Meta has introduced the Frontier AI Framework, outlining scenarios where it may stop developing or restrict access to advanced AI systems deemed too risky. These include AI models that could aid in cybersecurity breaches or biological attacks.

  • Why: The framework responds to concerns over Meta’s open AI approach. Unlike rivals like OpenAI, Meta has largely shared its AI models, but growing risks—such as Llama being used by U.S. adversaries—have prompted a more cautious stance.

  • Impact: Meta’s move signals growing industry awareness of AI dangers. While it seeks a balance between openness and safety, critics may question its ability to self-regulate without external oversight.

🩺 AI to speed up UK breast cancer screenings ↗️LINK

  • What: Nearly 700,000 women in the UK will be screened for breast cancer using AI as part of the Edith trial, a £11 million project designed to improve early detection and reduce NHS waiting times. AI will replace one of the two radiologists typically reviewing mammograms, potentially increasing efficiency and accuracy.

  • Why: The UK faces a radiologist shortage, leading to screening delays and late diagnoses. AI has shown promise in detecting breast cancer more accurately than doctors in smaller trials. If successful, this program could revolutionize cancer detection.

  • Impact: Faster, more accurate diagnoses could save lives while easing NHS burdens. If AI proves effective, a nationwide rollout could transform breast cancer screening, setting a precedent for AI use in broader healthcare diagnostics.

🔒 Anthropic dares hackers to break its AI ↗️LINK

  • What: Anthropic has unveiled Constitutional Classifiers, an AI safety system designed to block jailbreak attempts. It has withstood over 3,000 hours of bug bounty testing from 183 hackers, offering a $15,000 reward for a full breach—none succeeded. Public testing runs until February 10.

  • Why: As AI grows more powerful, preventing misuse is critical. Unlike traditional AI safety methods, this system generates diverse training data to detect and block manipulation attempts.

  • Impact: If effective, this approach could set a new standard for AI security, making AI models safer and more resistant to harmful exploits as they become more advanced.

🕶️ Meta’s smart glasses spending nears $100B ↗️LINK

  • What: Meta’s investment in augmented and virtual reality projects is set to exceed $100 billion by the end of 2025. Last year alone, the company poured nearly $20 billion into Reality Labs, which develops Ray-Ban Meta smart glasses and Quest VR headsets.

  • Why: CEO Mark Zuckerberg envisions 2025 as the defining year for smart glasses, aiming to build a computing platform that could replace smartphones and reduce Meta’s reliance on Apple and Google.

  • Impact: While the investment is staggering, success remains uncertain. Apple’s retreat from smart glasses raises doubts about market demand, but if Meta succeeds, it could reshape personal computing and lead the next tech revolution.

Top 5 AI Tools

  • ManyChat AI: Chat Marketing Made Easy with Manychat - Automate two-way, interactive conversations in Instagram Direct Messages, Facebook Messenger, and SMS to grow your brand.

  • InVideo AI: Create a video by typing simple text prompts like you would ask a human editor. Generate videos with powerful visuals, voices, and text using just one AI tool. realistic AI Voiceovers.

  • CapCut AI: AI video generator by CapCut can create stunning videos by simply entering your creative ideas. Let AI-generated videos boost your business or branding.

  • Kittl AI: Kittl is the most intuitive and easy-to-use design platform which helps you to create stunning designs that impress everyone. 

  • Trinka AI: Trinka AI is a privacy-first English grammar checker and language enhancement writing assistant designed for academic and technical writing.

- DO ME A FAVOUR -

If you find this email in your ‘Promotional or Spam’ tab, please move this email to your Primary Inbox.

I work so hard to bring all the latest AI news, tips, and tutorials directly to your inbox so that you don’t have to do the research by spending hours.

But if you don’t get to read my email, we both lose something.

I request you to move this email from the ‘Promotional Tab to Primary Inbox,’ so that you never miss my email and keep learning all the latest happenings in the AI Industry.

FEEDBACK

How would you rate this newsletter?

Your feedback is greatly appreciated and will help me improve future editions. Please take a moment to rate this newsletter on the scale below.

Login or Subscribe to participate in polls.

What did you think? We’re always looking for ways to improve. Reply with any feedback or interesting insights you might have!