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:

• AI Pioneers Hinton and Hopfield Win Nobel Prize

• Meta Ray-Bans Can Expose Strangers’ Personal Information with AI-Powered Doxing

• Uber to Launch GPT-4o Powered AI Assistant to Help Drivers Go Electric

• Meta Launches Llama 3.2 with Vision and Voice Capabilities

• And more AI news….

Top Developments

Nobel Prize

AI Pioneers Hinton and Hopfield Win Nobel Prize

The Gist: Geoffrey Hinton and John J. Hopfield snagged the 2024 Nobel Prize in Physics for their groundbreaking work in machine learning and artificial neural networks, shaping how we recognize images and analyze data.

Key Details:

• Hinton, known as the "Godfather of AI," was honored for his contributions to autonomous data analysis and image recognition.

• Hopfield's research focused on storing and reconstructing patterns in data, advancing our ability to work with complex information.

• Hinton has raised alarms about the potential risks of AI, especially as tech companies race to develop new models.

• After leaving Google in 2023, he is now focusing on discussing AI's dangers openly, emphasizing responsible development.

Meta

Meta Ray-Bans Can Expose Strangers’ Personal Information with AI-Powered Doxing

The Gist: Two Harvard students demonstrated how Meta’s Ray-Ban smart glasses could be used for AI-powered doxing, exposing personal information of strangers by leveraging facial recognition and public data, raising serious privacy concerns.

Key Details:

• The "I-XRAY" project uses Meta’s Ray-Ban glasses to stream video, identify faces, and pull publicly available personal data, such as addresses and even Social Security fragments.

• The project highlights the ease with which such technology can be used for harmful purposes, requiring only basic coding skills and public data.

• The developers, who worked on this as a privacy "wake-up call," stress that the tool is more about awareness than practicality, as any camera could perform similar tasks.

• Though the creators don’t plan to release their code, the project underscores the need for stronger data privacy protections in an increasingly AI-driven world.

Uber

Uber to Launch GPT-4o Powered AI Assistant to Help Drivers Go Electric

The Gist: Uber is introducing an AI assistant using OpenAI's GPT-4o to guide its drivers in transitioning to electric vehicles, aligning with the company's sustainability goals.

Key Details:

• The AI assistant will provide personalized EV transition advice based on location-specific incentives and driver preferences.

• Initially launching in the U.S. in early 2024, the chatbot will expand to other regions and broader queries by 2025.

• Uber has committed $800 million to help its driver partners fully transition to EVs by 2040.

• Nearly 180,000 drivers in North America and Europe are adopting EVs monthly, with 30% of Uber's miles in London now electric.

Quick Gist

Meta has launched Llama 3.2, its first multimodal open-source language model supporting vision and voice capabilities, with enhanced performance and deployment options for developers (Read More).

Google launched Gemini Nano on the Pixel 9 series, enabling on-device AI processing and multimodal capabilities, supported by MediaTek's upcoming Dimensity 9400 chipset (Read More).

Google is exploring the use of nuclear energy, including small modular reactors, to power its data centers and achieve net-zero emissions by 2030 amidst rising energy demands from AI (Read More).

Foxconn is constructing the world's largest production facility for Nvidia's GB200 superchips, marking a significant expansion into AI server technology (Read More).

Google is testing an AI button on the Android lock screen, possibly for future access to its Gemini features, although its functionality remains unclear for now (Read More).

Hugging Face launched the openai-gradio Python package, enabling developers to create AI-powered web apps quickly and easily with minimal coding effort (Read More).

ByteDance's web crawler, Bytespider, is rapidly outpacing competitors in data scraping, potentially to support the development of its own large language model as TikTok faces increased scrutiny from the U.S. government (Read More).

Vodafone and Google expanded their partnership for ten years to enhance mobile, broadband, and TV services with Google’s generative AI models, benefiting customers across Europe and Africa (Read More).

South Korean MLOps platform VESSL AI raised $12 million in Series A funding to optimize GPU costs and enhance AI model development through its hybrid infrastructure (Read More).

Google is launching new AI-powered security features for Android devices, including Theft Detection Lock and Remote Lock, to combat phone theft and protect user data (Read More).

Google's Gemini Nano, an AI model with multimodality, is currently exclusive to the Pixel 9 series but will soon be supported by MediaTek's upcoming Dimensity 9400 chipset, potentially debuting in Samsung's Galaxy S25 series (Read More).

Google DeepMind's implementation of its Gemini AI model aims to automate 40-50% of information requests in its legal department, emphasizing workflow efficiency and support for employees rather than job replacement (Read More).

Microsoft's Copilot Rewrite feature will integrate enterprise data protection for users with a Microsoft Entra ID, starting in February 2025, while also removing ads for a more seamless experience (Read More).

THAT’S IT FOR TODAY!