The #1 AI Newsletter for Business Leaders

Join 400,000+ executives and professionals who trust The AI Report for daily, practical AI updates.

Built for business—not engineers—this newsletter delivers expert prompts, real-world use cases, and decision-ready insights.

No hype. No jargon. Just results.

Subscribe free—trusted by leaders

Today:

• Elon Musk Reveals Grok 5 Master Plan with Colossus 2 Supercomputer

• Musk’s xAI Sues Apple & OpenAI Over AI App Store Bias

• Perplexity Launches $42.5M Revenue-Sharing Plan for Publishers

• From Humanoids to Smart Machines: Jetson Thor Powers Next-Gen Robots

• OpenAI Launches Learning Accelerator to Empower Students in India

Elon Musk GROK 5 Master Plan Revealed

Elon Musk’s xAI is building Colossus 2, a gigawatt-scale supercomputer aimed at training AGI. Grok 4, xAI’s flagship model, excels in reasoning thanks to 10x reinforcement learning (RL) compute. 

Despite being slower, it dominates key benchmarks and outperforms rivals on complex tasks. Musk believes edge-device AI, real-time software generation, and massive RL scaling signal a non-trivial path to AGI—potentially redefining how humans interact with software, cognition, and even reproduction.

WATCH THE VIDEO ON YOUTUBE

Musk’s xAI Sues Apple & OpenAI Over AI App Store Bias

Elon Musk’s xAI is suing Apple and OpenAI, claiming they teamed up to block competition in the AI space. The lawsuit says Apple favors OpenAI’s ChatGPT in its App Store, hurting xAI’s Grok and X apps. Musk argues this violates antitrust laws and stifles innovation. This legal battle adds to Musk’s ongoing fight against OpenAI’s shift to a for-profit model and Apple’s tightly controlled App Store practices.

Why it matters

1. Antitrust Tensions: The lawsuit challenges the growing dominance of AI partnerships, spotlighting concerns over monopolistic control by tech giants like Apple and OpenAI.
   

2. App Store Influence: It raises critical questions about platform neutrality and how app ranking can shape AI accessibility and market competition.
   

3. xAI vs. OpenAI Rivalry: Musk’s legal pressure highlights a larger war over the future of general-purpose AI and who controls its public distribution.

Perplexity Launches $42.5M Revenue-Sharing Plan for Publishers

Perplexity AI, a startup building an AI-powered search engine, is launching a $42.5 million revenue-sharing program for publishers. This move aims to ease tensions with media outlets who’ve criticized or sued over AI use of their content. The program offers media companies a share of the ad revenue their articles help generate. Perplexity hopes the plan will encourage collaboration rather than conflict as AI search becomes more common.

Why it matters

1. AI–Media Tensions
   It addresses growing legal and ethical concerns about AI systems using journalistic content without compensation.
   

2. New Monetization Model
   The initiative could establish a blueprint for how AI search engines and publishers can profitably coexist.
   

3. Search Engine Disruption
   By challenging Google and compensating creators, Perplexity is shaping a more sustainable, publisher-friendly AI search ecosystem.

From Humanoids to Smart Machines: Jetson Thor Powers Next-Gen Robots

NVIDIA has launched Jetson Thor, a powerful robotics computer built on Blackwell architecture, delivering real-time AI reasoning for physical robots. It boasts 7.5x more AI compute than its predecessor, enabling advanced edge processing for humanoids, smart machines, and multi-sensor AI agents. Partners like Agility Robotics and Boston Dynamics are integrating it into next-gen robots. With support for major AI models and frameworks, Jetson Thor marks a leap forward in physical AI.

Why it matters

1. Edge AI Breakthrough
   Jetson Thor brings server-level AI performance directly to robots, allowing fast, real-time decisions without relying on the cloud.
   

2. Agentic Robotics at Scale
   It empowers the next wave of agentic AI in robotics — from humanoids to surgical bots — with greater autonomy, memory, and multimodal reasoning.
   

3. Ecosystem Acceleration
   With support for major LLMs and vision-language-action models, Thor enables 2M+ developers to deploy powerful physical AI across industries faster than ever.

🧠RESEARCH

AgentFly: Fine-tuning LLM Agents without Fine-tuning LLMs

AgentFly introduces a new way to train AI agents without changing the core language model. It uses a smart memory system and reinforcement learning to help agents learn from experience. This method is cheaper, faster, and works well on real tasks, outperforming traditional fine-tuning in both accuracy and adaptability.

Beyond Pass@1: Self-Play with Variational Problem Synthesis Sustains RLVR

This paper improves how AI learns to solve tough problems by using a new self-play method. Instead of repeating the same examples, the AI creates slightly different problems with the same answers. This keeps the AI flexible, boosts accuracy across many guesses, and works well across multiple model sizes and benchmarks.

ODYSSEY: Open-World Quadrupeds Exploration and Manipulation for Long-Horizon Tasks

ODYSSEY is a new system that helps four-legged robots follow long instructions and handle real-world tasks. It combines language understanding, smart planning, and full-body control to work in messy environments. The system handles both rough terrain and precise object handling, proving useful for real-world indoor and outdoor challenges.

🛠️TOP TOOLS

Midjourney Prompt Generator - Designed to create detailed and effective prompts for the Midjourney AI image generation platform. 

BasedLabs AI - AI-powered tools for image and video generation

Mokker AI - AI-powered solution for instant, professional product photography, allowing users to generate high-quality images from a single product photo.

ChatGPT Sidebar - AI assistant for various tasks across web pages.

GhostCut - AI-powered video editing tool that offers advanced features such as automatic text removal, video translation, and dubbing in multiple languages.

📲SOCIAL MEDIA

🗞️MORE NEWS

• OpenAI launched the Learning Accelerator in India to support educators and students with AI tools like ChatGPT. It includes training programs, research with IIT Madras, and partnerships with schools to improve learning outcomes and digital skills.

• OpenCUA, a new open-source framework from HKU, enables the creation of AI agents that use computers like humans do. It outperforms open models and rivals OpenAI and Anthropic agents, offering scalable tools, datasets, and reasoning-based training.

• A blind test website lets users compare GPT-5 and GPT-4o without knowing which is which. Many still prefer GPT-4o’s warmth over GPT-5’s accuracy. The debate highlights how AI personality affects user trust.

• Google has upgraded its NotebookLM tool to support 80 languages for both Video and Audio Overviews. Originally English-only, this update now allows users to turn notes, PDFs, and images into video or audio summaries in their native language. Audio summaries are also more detailed for non-English speakers, with the option to choose short or full versions.