Welcome to Get The Gist, where every weekday we share an easy-to-read summary of the latest and greatest developments in AI—news, innovations, and trends—all delivered in under 5 minutes! ⏱

In today’s edition:

• Intel Leaves AI Race, Targets Cost-Effective Solutions

• X Allows Third-Parties to Train AI Models on User Data

• ChatGPT Launches Windows App

• Apple To Integrate ChatGPT into Siri

• And more AI news….

Top Developments

Intel

Intel Leaves AI Race, Targets Cost-Effective Solutions with Gaudi 3

The Gist: Intel is moving away from direct competition with NVIDIA in the AI hardware race, instead focusing on cost-efficient AI solutions with its Gaudi 3 accelerator, aimed at smaller businesses and startups.

Key Details:

• Intel's Gaudi 3 AI GPUs focus on performance-per-dollar value, offering 80% better value than NVIDIA’s H100 in inferencing tasks.

• Targeting smaller enterprises and startups, Intel is not pursuing mainstream market dominance like NVIDIA.

• Gaudi 3 has seen adoption by companies like IBM Cloud and Hewlett Packard Enterprise for data center products.

• Intel’s strategy focuses on cost-conscious AI solutions as the large-scale data center market shifts.

Twitter/X

X Allows Third-Parties to Train AI on User Data

The Gist: X (formerly Twitter) has updated its privacy policy to permit third-party collaborators to train AI models using user data, unless users explicitly opt out. The changes, effective November 15, open a new revenue stream for the platform.

Key Details:

• Third-party collaborators can use X user data to train AI models if users don’t opt out.

• Users will soon be able to toggle data-sharing in X’s privacy settings, though details are still unclear.

• X will retain user content and interactions for varying durations, potentially impacting AI data ingestion.

• A new clause imposes fees for scraping content, charging $15,000 per 1 million posts accessed within 24 hours.

OpenAI

ChatGPT Launches Windows App

The Gist: OpenAI has released a Windows app for ChatGPT, currently accessible only to paid subscribers. The app offers quicker access to the AI assistant but lacks some advanced features, such as voice mode.

Key Details:

• The app is available to ChatGPT Plus, Enterprise, Team, and Edu users via the Microsoft Store.

• Users can interact with ChatGPT using a dedicated window and the Alt + Space shortcut.

• It includes features like file uploads and access to OpenAI’s advanced o1 model.

• OpenAI plans to release the app to all users later this year.

Quick Gist

Adobe launched the Firefly Video Model, an AI tool that streamlines video production by enabling rapid content creation from text prompts, reshaping corporate media strategies and reducing costs (Read More).

Google updated NotebookLM with a "Guide the Conversation" feature for customized AI-generated audio overviews, enhancing user control and experience while maintaining data privacy (Read More).

Anthropic updated its Claude AI mobile apps with new features for personalized instructions and the introduction of an iPad app, enhancing user experience and productivity (Read More).

The Open Source Initiative criticized Meta Platforms for misleadingly promoting its Llama AI models as open source, arguing it undermines the definition of open source and lacks necessary transparency (Read More).

Researchers at the University of Zurich developed an AI system based on GPT-4 to enhance the detection of antibiotic-resistant bacteria, aiming to improve diagnostic speed and accuracy in healthcare (Read More).

Meta unveiled new AI hardware designs at the OCP Global Summit 2024, including the Catalina rack and advanced networking solutions, to enhance its AI infrastructure and support its massive Llama 3.1 model (Read More).

The Ministry of Skill Development & Entrepreneurship in India partnered with Meta to launch an AI Assistant for the Skill India Mission and establish five Centers of Excellence in Virtual Reality and Mixed Reality to enhance skill training for the youth (Read More).

Meta is extending its pilot of the generative AI video model Movie Gen through 2025 in collaboration with Blumhouse, enabling filmmakers to create custom videos and sounds from text inputs (Read More).

Google is transferring its Gemini app team to DeepMind to streamline operations and enhance its generative AI capabilities (Read More).

Apple is integrating ChatGPT into Siri with new Apple Intelligence features, set to debut in iOS 18.2 and macOS 15.2 updates by December 2024 (Read More).

THAT’S IT FOR TODAY!