- AI PlanetX
- Posts
- Dario Amodei Fears China’s AI Power
Dario Amodei Fears China’s AI Power
Microsoft Unveils Cybersecurity Model

Welcome to another edition of AI PlanetX.
Anthropic warns of China AI risks, Microsoft launches AI cybersecurity tools, and Cursor expands into India with its first local subscription.
Inside This Edition: 💎
Hottest AI News
Top AI & SaaS Tools
AI Tutorial: TikTok-Style Videos with Notebook
Top AI & Tech News
AI Art Spotlight
Prompt of the Day: NotebookLM Study Guide Prompt
Featured AI Video
Course of the Day: Hermes Agent Full Course
Hottest AI News
Anthropic
Anthropic Rejects Open-Weight Bans but Warns of Serious China AI Risks

Anthropic CEO Dario Amodei says he does not support banning open-weight AI. His concern is narrower: powerful Chinese models, authoritarian misuse, and risks that cannot be fully reversed once model weights are released.
Details:
His reply followed an Nvidia-led open letter from Meta, Microsoft, Hugging Face, Mistral and others warning against broad restrictions. Amodei called non-dangerous open models a public good
His fear is that China or other authoritarian states could gain military power, repress people, or enable cyber and biological attacks. Open weights are harder to guard, monitor or withdraw
He supports chip controls, distillation crackdowns and sanctions for proven IP theft, plus global safety test for the strongest models, open or closed, while exempting startups
Open-source advocates say AI availability helps defenders. Amodei says global cooperation, including China, may still prevent AI-enabled biological weapons.
The Ultimate Claude Code Guide to ship like Anthropic engineers
AI will write 90% of code by the end of 2026, and only 10% of developers will stay relevant.
The engineers in that 10% aren't smarter. They just know how to use AI.
We put together the exact playbook Anthropic engineers use to make sure you're in the top 10%.
Sign up for The Code and get access to:
The Ultimate Claude Code Guide 2026 — 50+ tips and tricks to code 5x faster
The Code newsletter — learn the latest AI tools, tips, and skills to code faster with AI in 5 minutes a day
Microsoft
Microsoft Launches Its First AI Cybersecurity Model and Agent Platform

Microsoft is entering the AI cybersecurity race with MAI-Cyber-1-Flash and Project Perception, a model-and-agent stack built to find, prioritize and fix software vulnerabilities at machine speed.
Details:
MAI-Cyber-1-Flash runs inside MDASH with GPT-5.4. Microsoft says it scored 96% on CyberGym, beating Mythos, Gemini and GPT, while cutting costs by about 50%
Perception uses red agents to simulate attacks, blue agents to investigate and prioritize risks, and green agents to apply fixes—turning hours of specialist work into quicker defense
The race is growing: Anthropic’s restricted Mythos 5 handles advanced cyber work, Fable 5 adds stronger safeguards, and OpenAI’s Daybreak supports approved defenders
Microsoft says Perception enters public preview August 3, as AI security moves from finding bugs to continuously detecting, fixing and hardening code.
Top AI & SaaS Tools
Acadle (Life-time Deal): AI-powered platform for creating and managing online academies with automated course generation, assessments, certificates, and performance tracking
BUZZ: Open-source, decentralized, self-sovereign, model-agnostic group chat platform for teams and agents built to reduce dependence on Slack and GitHub
Cekura: Tests voice and chat AI agents at scale, finds and fixes failures by updating prompts/configs, and re-validates with full regression to ensure the fix holds
Vida: Proactive AI desktop assistant that learns your work context across apps, anticipates what you need, and helps complete tasks [F-R-E-E to Try]
RevPDF: Offline, cross-platform PDF editor offering full editing, annotation, signing, and conversion features without requiring internet or accounts [F-R-E-E]
Own Search With Podcasts
Search engines and AI platforms reward brands that are mentioned, cited, and trusted across the web. PodPitch books your experts on relevant shows, creating branded mentions, backlinks, transcripts, citations, and reusable content. Only 20 demo spots are available this month. Once claimed, the offer disappears.
AI Tutorial
How to Create TikTok-Style Short Videos with Gemini Notebook

Creating short educational videos from research no longer has to start with a blank script. NotebookLM can read your uploaded sources, understand the main ideas, and help you generate a short visual overview from them.
Sign in to NotebookLM
Go to NotebookLM and sign in with your Google account.
Create Notebook
Choose the notebook that already has your research, or create a fresh notebook for this topic.
Upload Sources
Add the materials you want NotebookLM to use. You can upload things like:
PDFs
Google Docs
Website links
Notes or pasted text
Go to Studio
Head to the Studio panel on the right side of NotebookLM, choose Video Overview, then select the Short format to turn your research into a quick, TikTok-style video.
Pick Focus Topic
Choose one of NotebookLM’s suggested topics, or type your own custom focus. For example:
“Explain the main idea for beginners”
“Turn this research into a short social media explainer”
“Focus on the most surprising findings”
Generate Clip
Tap Generate and let NotebookLM create the video. Once it is ready, review the clip and make sure the main points match your research.
Note: Use it for repurposing, not replacing. Your own hook, caption, and final editing can still make the clip much stronger.
Top AI & Tech News
Cursor is making a major push into India by launching its first country-specific subscription, Cursor Start, priced at ₹649/month
Meta and BlackRock are partnering on a $14 billion El Paso data center venture, with BlackRock’s funds owning 80% and Meta leasing the 1-gigawatt AI campus
Canadian politician Bill Oliver went viral after seemingly reading AI-generated editing notes aloud during a speech to lawmakers, accidentally doing it twice
A Reddit thread said a simple Google search exposed thousands of publicly shared Claude conversations, prompting a privacy backlash and Anthropic to remove the indexed chats
AI Art Spotlight

Create Your Own Photo Like This Using the Prompt Below:
[subject], premium food advertising, white seamless background, high key studio lighting, floating stacked composition, levitating pieces, [ingredient_bits] scattered, clean minimal layout, crisp soft shadow on surface, ultra realistic, macro product photography, 100mm lens look, f/8 sharp focus, subtle texture detail, editorial commercial, 8kModel: Nano Banana 2
Prompt of the Day
The Ultimate NotebookLM Study Guide Prompt

This prompt turns your NotebookLM sources into a personalized study guide based on your topic, knowledge level, and learning goals. It organizes the material into explanations, key concepts, source summaries, examples, practice questions, exam tips, visuals, and a revision checklist.
Featured AI Video
Fine-Tune Any AI Model Locally (Full Tutorial)
AI Course of the Day
Hermes Agent Full Course: 3 Hours — Build & Sell (2026)

This course shows how to start an AI consulting business for small business owners, even without coding skills, money, or an audience. Corey Gannon explains a simple 45-minute “AI tools assessment” to find bottlenecks and recommend off-the-shelf AI tools that can save hours each week.


