📅 November 2024 Summary

🦃 if you haven't kept up on AI, you need to read this email

Henry Habib
November 30, 2024

Subscribe | Sponsor

Hello folks,

Special issue here: a collection of all the stories, news, and cool AI art you missed in August.

Let’s get right into it.

In this issue:

🤝 In Partnership: Writer, the easiest way to build apps
📰 Top News: Summary of top AI, no-code, automation news
🖼️ AI Art: Summary of top AI Art

🤝 IN PARTNERSHIP WITH WRITER

Writer RAG tool: build production-ready RAG apps in minutes

RAG in just a few lines of code? We’ve launched a predefined RAG tool on our developer platform, making it easy to bring your data into a Knowledge Graph and interact with it with AI. With a single API call, writer LLMs will intelligently call the RAG tool to chat with your data.

Integrated into Writer’s full-stack platform, it eliminates the need for complex vendor RAG setups, making it quick to build scalable, highly accurate AI workflows just by passing a graph ID of your data as a parameter to your RAG tool.

Learn more about our production ready RAG tooling here.

📰 TOP NEWS

Summary

1/ An AI-generated fake news story about a Dublin Halloween parade caused chaos when thousands gathered for the non-existent event. Shared widely via SEO and TikTok, the hoax disrupted transport and required police intervention.

2/ Amazon Prime Video introduces X-Ray Recaps, an AI tool that generates concise summaries for shows, episodes, and scenes, enhancing viewer comprehension while avoiding spoilers.

3/ Microsoft unveils Magentic-One, an open-source multi-agent AI system for task management. Powered by an Orchestrator agent and four specialized agents, it handles tasks like financial analysis, citation management, and food ordering. Designed with GPT-4o but model-agnostic, it boosts productivity and supports commercial use.

4/ Physical Intelligence, a San Francisco startup backed by Jeff Bezos and OpenAI, has raised $400M to develop household robots. Using AI models like pi-zero, these robots adapt to tasks such as laundry, table bussing, and food packing, though challenges in safety and self-improvement remain.

5/ AI-powered high-speed video may enable non-invasive screenings for high blood pressure and diabetes, achieving up to 94% accuracy in detecting hypertension. The system analyzes blood flow from facial and hand images. While promising, further validation and accuracy improvements are needed for widespread use.

6/ Anticipation grows for Google’s Gemini 2.0, with leaks hinting at a powerful AI upgrade amid rivalry with OpenAI. Reports suggest Gemini’s experimental model outperformed OpenAI’s latest tests, but no release date has been confirmed.

7/ OpenAI's ChatGPT Windows app is now available to all users, expanding access beyond its initial release for paid users only.

8/ Virgin Media O2's AI chatbot, Daisy, mimics an elderly woman to outsmart scammers through endless chatter, disrupting their efforts and aiding law enforcement.

9/ Microsoft integrates Azure SQL into Fabric, creating an AI-ready data layer for seamless access to live and historical data. This move enhances enterprise AI workflows, supports next-gen applications like ChatGPT, and simplifies data interaction with natural language through new AI Skills.

10/ Microsoft is leading the AI agent era, with 100,000 organizations adopting AI agents. The shift in enterprise AI value from LLMs to governance layers highlights Microsoft's strategic advantage. Their vision of a multi-agent mesh could transform AI architecture, but startups worry that Microsoft's dominance may disrupt entire industries.

11/ Claude AI, developed by Anthropic, enhances productivity with a new feature that allows users to analyze Google Docs within chats and projects. Available to Claude Pro and Claude for Work users, it extracts text from documents to provide context in conversations, syncing directly from Google Drive for real-time updates.

12/ Startup Pickle has launched an AI avatar service that lets users create a virtual presence for web conference calls. By submitting a five-minute training video, users can have a personalized avatar ready in a day. The service supports Zoom, Google Meet, and Teams, with subscription tiers ranging from $300 to $1,150 annually. Currently available on MacOS, a Windows version is coming next month.

13/ NVIDIA has unveiled Fugatto, a generative AI model that creates and transforms music, voices, and sounds from text and audio inputs. Offering versatile control, it allows users to modify songs and voices and even create new sounds. Fugatto's unique ability to handle untrained tasks, such as generating high-quality singing voices, opens up applications in music, advertising, language learning, and gaming. It uses ComposableART for fine-tuned text instruction control.

🖼️ AI ART

Examples of great and trending AI art