- Weekly AI News
- Posts
- OpenAI Sora leak 🕵️♂️, Anthropic Custom Styles 🎨, Luma AI Updated ✨
OpenAI Sora leak 🕵️♂️, Anthropic Custom Styles 🎨, Luma AI Updated ✨
OpenAI's text-to-video AI tool, Sora, was leaked on the developer platform Hugging Face by artists protesting the company's exploitation of unpaid labor during its testing phase, Anthropic has launched custom styles for Claude.ai, allowing users to personalize AI responses, & More...
Transform Your Excel Skills with 50 Pro Hacks (+ Free Templates)
Stop wrestling with spreadsheets. Our comprehensive Excel mastery pack includes everything you need to become a spreadsheet wizard in record time.
Master 50 essential Excel tricks with step-by-step GIF tutorials
Create stunning data visualizations using done-for-you graph templates
Learn time-saving shortcuts the pros use daily
Access game-changing formulas and functions explained simply
Ready to revolutionize your Excel workflow?
💬 OpenAI Developments
OpenAI’s text-to-video AI tool Sora leaked in protest by artists
OpenAI's text-to-video AI tool, Sora, was leaked on the developer platform Hugging Face by artists protesting the company's exploitation of unpaid labor during its testing phase. The artists accused OpenAI of prioritizing public relations over meaningful collaboration, sparking concerns about the tool's impact on creative industries, including potential job losses in film.
OpenAI gets new $1.5 billion investment from SoftBank, allowing employees to sell shares in a tender offer
SoftBank is set to invest up to $1.5 billion in OpenAI through a tender offer, allowing current and former employees to sell their shares. This move aims to bolster SoftBank's position in the AI sector, following its previous $500 million investment in OpenAI's October funding round, which valued the company at $157 billion.
🌐 AI in Various Sectors
Healthcare: Hippocratic AI has been granted U.S. Patent No. 12,142,371 for its innovative constellation architecture, which enhances the safety of large language models (LLMs) in healthcare applications. This architecture employs a primary model for patient interactions, supported by multiple specialized models—such as the Overdose Engine and Medication Reconciliation Engine—to ensure accurate and safe responses in clinical settings.
Robotics: Researchers from the University of Pennsylvania have demonstrated that AI-enhanced robotic systems are highly susceptible to jailbreaks, enabling malicious commands such as instructing robot dogs to attack humans, directing bombs to cause maximum damage, and programming self-driving cars to intentionally harm pedestrians. Their study, which achieved a 100% success rate in compromising systems like Nvidia's Dolphins LLM, Clearpath Robotics' Jackal UGV, and Unitree Robotics' Go2 quadruped, underscores the urgent need for robust security measures in AI-integrated robotics to prevent potential real-world dangers.
Social Media and Interaction: Rox, a startup aiming to revolutionize the CRM market, is building an AI-native platform to challenge Salesforce by focusing on revenue teams. Unlike traditional CRMs, Rox leverages AI to integrate seamlessly with companies’ existing data warehouses, offering real-time, personalized insights through "agent swarms" that streamline sales processes. Rox’s strategy focuses on delivering immediate value with AI tools, gradually embedding deeper into workflows to become the system of record for revenue. With $50 million in funding and a long-term vision of automating revenue operations, Rox positions itself to thrive as AI capabilities advance.
🚀 Tech Industry Moves
Nvidia has introduced Fugatto, an AI audio model capable of generating novel sounds, such as a trumpet mimicking a dog's bark, and transforming existing audio by altering instruments or vocal characteristics. While Fugatto showcases advanced capabilities in sound generation and manipulation, Nvidia has not yet announced plans for its public release, citing concerns over potential misuse.
Microsoft has clarified that it does not use customer data from Microsoft 365 apps, such as Word and Excel, to train its AI models. This announcement addresses recent online rumors suggesting users needed to opt out to prevent their data from being used for AI training.
Microsoft has introduced the Zero Day Quest, a bug bounty program offering $4 million in rewards for identifying vulnerabilities in its cloud and AI technologies. This initiative aims to enhance security by fostering collaboration between Microsoft engineers and external researchers, with top contributors invited to an in-person hacking event at Microsoft's Redmond headquarters in 2025.
Google: In the ongoing antitrust trial, U.S. District Judge Amit Mehta indicated that artificial intelligence (AI) will play a significant role in determining remedies for Google's search monopoly, acknowledging AI's potential to reshape the search market. In response, Google is seeking information from AI competitors, including details of Microsoft's $13 billion investment in OpenAI, to argue that proposed remedies could hinder its ability to compete in the evolving AI landscape.
Anthropic has introduced the Model Context Protocol (MCP), an open-source standard designed to seamlessly connect AI assistants with various data sources, including content repositories, business tools, and development environments. By providing a universal protocol, MCP simplifies the integration process, enabling AI systems to access and utilize data more effectively, thereby enhancing their performance and relevance.
💡 Emerging AI Technologies and Companies
Paris-based Argil raises €4.9 million to develop AI video creation for the creator economy
Paris-based AI startup Argil has secured €4.9 million in funding, led by EQT Ventures, to enhance its platform that enables creators, educators, and businesses to produce high-quality videos using hyper-realistic avatars. Emerging from the Y Combinator incubator, Argil's technology allows users to generate multilingual, ready-to-post videos in minutes, aiming to democratize video production by making it as simple as writing a tweet.
Luma AI updates Dream Machine with new Photon image generation model and more
Luma AI has introduced significant updates to its Dream Machine platform, including the high-performance Photon Image Model for faster, higher-quality image generation, a revamped user interface for intuitive creation, and new subscription tiers tailored to hobbyists, professionals, and enterprises. Dream Machine now supports advanced prompts, image-based inputs, and collaboration tools, enabling creators and businesses to produce detailed visuals efficiently. The platform's new API also allows developers to integrate its capabilities into third-party apps while ensuring privacy. With over 25 million users, Luma AI aims to redefine visual creativity and storytelling for diverse industries.
🎨 AI in Creative Industries
Anthropic has launched custom styles for Claude.ai, allowing users to personalize AI responses to suit their unique communication needs and workflows. Users can select preset styles like Formal, Concise, or Explanatory, or create custom styles by uploading sample content and providing specific instructions. These features enable Claude to adapt its tone and structure for various applications, from technical documentation to marketing. Companies like GitLab are already leveraging these styles to enhance consistency and efficiency across tasks. Users can easily start customizing styles directly in the Claude.ai chat interface.
🌍 Global AI Race
South Korea Becomes Taiwan’s Largest Trade Deficit Due to NVIDIA AI Supply Chain
In the first ten months of 2024, South Korea became Taiwan's largest source of trade deficit, reaching a record $18.1 billion, primarily due to imports of integrated circuits. This deficit is largely attributed to South Korea's supply of DRAM to Taiwanese firms for packaging and integration, as both nations collaborate in the AI supply chain for companies like NVIDIA.
If you want more updates related to AI, subscribe to our Newsletter
👥 Connect & Feedback!
👉 Join Us:
📧 Advertise In Weekly AI News:
📧 Contact directly at [email protected]
😍 Share with your friends!

Reply