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:

• Gemini AI Delayed for iPhones Amid Apple's ChatGPT Partnership

• Google Docs Now Lets You Create AI-Generated Images

• TwinMind Launches ChatGPT Competitor

• Google’s Gemini AI Asks Student to Die

• And more AI news….

Top Developments

Gemini

Gemini AI Delayed for iPhones

The Gist: Google's Gemini AI won’t integrate with iPhones until at least 2025 as Apple prioritizes ChatGPT for its AI suite, raising questions about the exclusivity of Apple’s partnership with OpenAI.

Key Details:

• Apple's AI suite, "Apple Intelligence," will launch globally in December 2024 with ChatGPT but exclude Europe and China initially.

• Integration of Google’s Gemini AI with iOS is now expected no earlier than 2025, possibly aligning with the iPhone 17 release.

• Apple hinted at collaborating with multiple AI platforms at WWDC 2024 but appears to focus solely on ChatGPT for now.

• The delay has sparked speculation about whether the decision is a strategic move or part of a contractual arrangement with OpenAI.

Google

Google Docs Now Lets You Create AI-Generated Images

The Gist: Google Docs users can now generate in-line AI images directly in their documents using the Imagen 3 model, accessible via the Gemini sidebar.

Key Details:

• The new "Create an Image" feature is available under the insert menu, where users can describe the image, select an aspect ratio, and choose a style like photography or watercolor.

• Workspace subscribers currently have access, with Gemini Advanced subscribers gaining the feature in the coming weeks.

• Users can view multiple iterations of generated images and select their preferred version for the document.

• Google plans to enhance Gemini with upcoming versions, including the Gemini 2.0 model and "Project Jarvis," an AI agent for automating web-based tasks like research and bookings.

TwinMind

TwinMind Launches a Personal Jarvis for Your Smartphone

The Gist: Startup TwinMind, founded by former Google X employees, unveils an AI app that constantly listens, remembers, and learns about its users to offer hyper-personalized assistance akin to Tony Stark’s Jarvis.

Key Details:

• TwinMind runs in the background, transcribes audio to text, and uses this information to provide tailored insights and assistance, including summaries of daily activities.

• The app integrates with calendars, Gmail, and a browser extension to enhance productivity and memory-driven task execution, such as drafting emails or understanding past interactions.

• TwinMind processes most data locally on devices, minimizing privacy concerns, and claims 12-hour background operation without excessive battery drain.

• Available for iPhone and Chrome initially, TwinMind is free with a $20 monthly premium tier for advanced features, with Android and Mac apps arriving in 2025.

Quick Gist

NASA and Microsoft's 'Earth Copilot' offers an AI-powered interface for analyzing NASA's datasets via natural language queries, supporting NASA’s Open Science initiative and promoting accessibility to complex Earth science data (Read More).

Nvidia's Blackwell AI chips are experiencing overheating issues in server racks, causing delays and raising customer concerns about operational timelines and the chips' reliability (Read More).

Elon Musk's lawsuit against OpenAI now includes Microsoft and antitrust claims, alleging that the companies are engaging in anti-competitive practices and favoring their own platforms (Read More).

OpenAI's upcoming 'Live Camera' feature for ChatGPT’s Advanced Voice Mode aims to improve real-time visual capabilities, currently in beta testing (Read More).

Virgin Media O2's "Daisy" AI chatbot mimics lifelike conversations to engage phone scammers, effectively wasting their time and protecting users from fraud (Read More).

Google's Gemini AI chatbot faced backlash after delivering a threatening message to a student, sparking concerns about the accountability and safety of generative AI systems (Read More).

THAT’S IT FOR TODAY!