Today:

• Musk's xAI Secures $6 Billion Funding

• Sam Altman, Tech CEOs Join AI Safety Board

• Meta Sheds $200 Billion After AI Investment Update

• Samsung, Google Partnership

• AI-powered Ads on Reddit

• Edia's AI-Powered Math

• AI Tupac Voice Sparks Legal Battle for Drake

Musk’s xAI Is Close to Raising $6 Billion from Sequoia, Others

Elon Musk's xAI is on the verge of securing $6 billion from investors like Sequoia, aiming to boost its chatbot Grok's capabilities to rival OpenAI's ChatGPT. The year-old startup's funding round values it at $18 billion, excluding the investment. This move signals fierce competition among startups developing human-like technology for writing, conversation, and art creation. 

With Sequoia Capital among the backers, the deal is set to close within two weeks. xAI's push for funds underscores the intense race in the artificial intelligence sector, with players like OpenAI and Anthropic vying for dominance in mimicking human capabilities.

THE INFORMATION

OpenAI’s Sam Altman and Other Tech Leaders to Serve on AI Safety Board

Sam Altman, alongside CEOs from Nvidia, Microsoft, and Alphabet, will serve on a new federal advisory board focused on ensuring safe AI deployment in critical US infrastructure. The Artificial Intelligence Safety and Security Board aims to protect the economy and vital industries from potential AI-related threats. 

This move reflects the Biden administration's response to the regulatory gap surrounding the rapid advancement of AI technology. The initiative underscores the importance of safeguarding against potential risks associated with AI in critical infrastructure.

THE WALL STREET JOURNAL

Meta loses nearly $200 billion in value after Mark Zuckerberg announcement

Meta, formerly Facebook, faced a stock plummet of nearly $200 billion after CEO Mark Zuckerberg revealed slower-than-expected progress in AI investment. Despite Zuckerberg's reassurances about future revenue streams from AI products, investors remain skeptical, leading to a 15% drop in Meta shares. This downturn also impacted other tech giants like Snap, Alphabet, Amazon, and Microsoft. 

Zuckerberg outlined various monetization strategies for AI, including business messaging, ads, and paid content. However, the company's immediate financial outlook remains uncertain, sparking concerns about its long-term viability in the market.

INDEPENDENT

Samsung and Google tease “exciting things” for AI on Android and Galaxy hardware

Samsung and Google hint at upcoming AI developments for Android and Galaxy devices, emphasizing a strengthened partnership. This announcement follows Google's merger of its Android and hardware teams. Expect increased AI integration beyond features like Circle to Search. The Rabbit R1 represents a growing trend of AI-powered devices entering the market. 

Comments reflect mixed reactions, with some questioning Samsung's status and others critiquing its UX and voice assistant. Amidst discussions, topics like SD card support and the evolution of storage as a service arise. Overall, anticipation builds for the future of AI in mobile technology.

 THE VERGE

Reddit, AI spam bots explore new ways to show ads in your feed

Reddit introduces Dynamic Product Ads, utilizing machine learning and user engagement data to display relevant ads. This move emphasizes Reddit's shift towards an ad-centric model, with ads comprising 98% of its revenue. The platform aims to leverage user discussions for targeted advertising while enhancing the shopping experience. 

Despite user concerns about increased ad presence, Reddit defends its strategy, stating that its communities naturally engage in commercial activities. With plans for further ad integrations and merchant platform collaborations, Reddit solidifies its commitment to advertising as its primary revenue stream. 

ARS TECHNICA

AI startup Edia guarantees better math outcomes for school districts within one year

Edia, an AI-powered math education platform, promises school districts improved student math performance within a year or a full refund. Its solution offers personalized feedback for students, real-time data for teachers, and targeted interventions. The company secured a $9.4M Series A funding led by Felicis, signaling confidence in its approach. 

Edia stands out with features like allowing students to type out math on Chromebooks effortlessly and AI-guided step-by-step assistance. Its predictive analytics help teachers identify student needs promptly, contrasting with traditional quarterly assessments. Despite competition from established players, Edia aims to revolutionize K-12 education with its modern, teacher-centric platform.

VENTUREBEAT

Drake threatened with lawsuit over diss track featuring AI Tupac

Tupac Shakur's estate is threatening legal action against Drake for using an AI-generated Tupac voice on his diss track "Taylor Made" without permission. Drake's recent track, part of ongoing rap beef, features AI-created voices of Tupac and Snoop Dogg. Tupac's legal team calls it a "blatant abuse" of his legacy. 

The feud between Drake and Kendrick Lamar led to this release, drawing attention to the growing trend of AI-generated music. Despite Drake's history of criticizing AI music, his use of it in "Taylor Made" has sparked controversy and potential legal consequences.

THE VERGE

🧠RESEARCH

MaGGIe: Masked Guided Gradual Human Instance Matting

MaGGIe is a new approach to human matting in images and videos. It gradually predicts alpha mattes for each human instance, maintaining precision and consistency without increasing computational costs. MaGGIe utilizes transformer attention and sparse convolution, achieving robust performance across various scenarios, enhancing model generalization in real-world applications.

XC-Cache: Cross-Attending to Cached Context for Efficient LLM Inference

XC-Cache is a method for efficient language model inference. It addresses the inefficiency of in-context learning by using cross-attention to condition generation on reference text without prompts. By leveraging pre-trained models and adding few layers, it outperforms traditional methods, drastically reducing space requirements while maintaining performance in conditional generation tasks like Question-Answering.

MultiBooth: Towards Generating All Your Concepts in an Image from Text

MultiBooth is a method for efficient multi-concept customization in image generation from text. It addresses challenges in fidelity and inference cost by dividing the process into single-concept learning and multi-concept integration phases. By using bounding boxes and cross-attention maps, it achieves improved concept fidelity and computational efficiency, outperforming baselines in evaluations.

Music Consistency Models

The paper introduces Music Consistency Models (MusicCM) for efficient music generation. Inspired by consistency models in image/video generation, MusicCM synthesizes mel-spectrograms for music clips with minimal sampling steps, enhancing computational efficiency while maintaining high quality. By incorporating consistency distillation and multiple diffusion processes, it achieves seamless music synthesis in real-time applications.

SEED-X: Multimodal Models with Unified Multi-granularity Comprehension and Generation

SEED-X is a multimodal foundation model aimed at enhancing vision-language understanding and generation. Addressing limitations in real-world applicability, it focuses on comprehending images of varying sizes and ratios and enabling multi-granularity image generation. SEED-X demonstrates competitive performance on benchmarks and effectiveness in real-world applications across domains, inspiring future research in multimodal models.

🛠️TOP TOOLS

Micro1 - The first ever AI interviewer

Muddy - The browser that knows you and your crew

Zapier - Create an AI chatbot that answers questions, resolves issues, and nurtures leads with the power of automation.

Weavely- Design and publish interactive, web-based forms by transforming your visual ideas with Weavely, a Figma-based form design tool

Spiritme - AI-video platform for creating personalized videos with digital avatars. 

🗞️MORE NEWS

OpenAI's Chris Lehane says AI is "critical infrastructure"

OpenAI's Chris Lehane positions AI as crucial infrastructure, paralleling the New Deal's impact on America. Lehane, OpenAI's new VP of public works, stresses AI's potential in healthcare, education, and climate solutions, aligning its success with societal progress. OpenAI aims for broader engagement beyond CEO Sam Altman, acknowledging the need for diverse voices. This shift responds to public skepticism and aims to showcase AI's tangible benefits. As OpenAI navigates its growth, challenges like regulation and public perception loom, requiring strategic communication and tangible results to win public trust.  AXIOS

New AI technology estimates brain age using low-cost EEG device

Researchers at Drexel University have developed a new artificial intelligence technique that estimates an individual's brain age based on electroencephalogram (EEG) brain scans. This technology could facilitate early screening for degenerative brain diseases by making brain-age estimation more accessible. The method utilizes machine learning to analyze EEG data and estimate brain age, similar to how one might guess a person's age based on physical appearance. The technology has been licensed to DiagnaMed Holdings for integration into a new digital health platform, potentially enhancing brain health monitoring and optimization strategies. MEDICAL XPRESS

China's mega-telcos are spending billions on AI servers

China's major telecom companies are investing billions in AI servers, with China Mobile seeking nearly 8,000 machines by 2025. This massive procurement, likely exceeding $2 billion, marks the largest centralized purchase of AI servers in China. While the exact purpose remains undisclosed, possibilities include cloud operations or customer service enhancement through generative AI. THE REGISTER

AI startup Prime Intellect raises $5.5M to build high-powered, decentralized research platform

Prime Intellect, a startup, has secured $5.5 million in funding to launch a decentralized platform for collaborative AI research. The platform aims to bridge the gap between academic research and private-sector ventures by providing a space for researchers to collaborate and build open-source AI models. By aggregating global compute power, Prime Intellect intends to make large-scale AI development more accessible and not reliant on Big Tech. The platform will operate as a marketplace for GPUs, enabling users to contribute compute power, code, and capital to collectively train AI models. FORTUNE

Elon Musk's Big Bet: Tesla to Ramp Up AI Development with Nvidia's Powerful Chips

Elon Musk plans to boost Tesla's AI development by installing 85,000 Nvidia chips by year-end. This move could make Tesla a major Nvidia customer, surpassing tech giants like Amazon and Google. Musk aims to advance AI for projects like Optimus Bot and Dojo supercomputer. Despite Nvidia's recent stock drop, analysts remain bullish on its future. YAHOO!