Samsung Gauss - a new AI on-device model introduced

Today's issue is brought to you by our sponsor - Koala. Accelerate Your Content Creation with this (better than ChatGPT) AI Tool!

📰 No-Noise News 📰

🤖 Samsung Research has introduced an AI model - Samsung Gauss. Structured as three tools: Gauss Language, Gauss Code, and Gauss Image, it can understand languages, write code, and enhance images respectively.

🦾 Amazon is developing a new AI model, Olympus, that rivals ChatGPT and Bard. This move aims to boost Amazon's competitiveness in the AI market and should enhance the customer experience in their online store and Alexa services.

🎨 Microsoft is rolling out a service that utilizes digital watermarks to denote AI-generated content. This aims to curb the spread of deepfakes and bolster cybersecurity for politicians, particularly ahead of various elections worldwide.

🧑🏿‍🤝‍🧑🏿 The actors' union, SAG-AFTRA, has rejected an AI clause proposed by AMPTP, which aims to capture and use actors' likenesses without consent or compensation after their death. This remains one of the sticking points prolonging the actors' strike.

⚙️ Amid an AI chip shortage, Microsoft is partnering with venture capitalists, allowing select startups to utilize its Azure cloud "supercomputing" resources at zero cost. It's a strategic move to drive AI innovation.

💸 IBM, a contributor to Hugging Face's Series D funding, is deepening its commitment to AI, announcing a $500M enterprise AI fund. The venture will support a variety of AI businesses, emphasizing generative AI technology for large enterprises.

🎙️ Barack Obama on AI, free speech, and the future of the internet - Obama outlines the need for comprehensive AI regulation that protects public interests, promotes transparency, and encourages tech experts to contribute to governmental oversight for positive societal impact.

📚 Nerd section 📚

⚒️ New method reveals how one LLM can be used to jailbreak another - Penn students have developed an algorithm, named PAIR, that can auto-detect safety vulnerabilities in large language models (LLMs). It's built to spot potential "jailbreak" prompts that can cause these LLMs to produce harmful content. Interestingly, it can be utilized with black-box models like ChatGPT, where it successfully generates potential threats with fewer attempts.

🎞️ The Fraunhofer Heinrich Hertz Institute has introduced a new method using Neural Radiance Fields (NeRFs) for creating animated 3D models from regular video footage. This could bypass the need for expensive equipment, making the process more accessible and less resource-intensive.

🌐 Web section 🌐 

So here (but not in the email version) we're adding AI-related stories that may be slightly less attention-grabbing, but still could move the needle in specific sectors and fields.

🎶 AI can narrate sports matches with ChatGPT and TTS - while the technology offers a compelling proof of concept, the lack of human-like excitement in the commentary indicates there's room for improvement.

🤖 Humans vs. robots: Study compares 27 humanoid robots with humans to see who is superior. Spoiler: overall, we are (still) superior in executing multi-purpose tasks and being energy efficient.

😏AI-generated faces look just like real ones – but evidence shows your brain can tell the difference - studies show that our brain instinctively picks up on minute differences, demonstrating an unconscious awareness.

🤯  Microsoft is using AI tools to enhance moderation in Xbox. These tools automate content flagging for human review, minimizing bad user content and enhancing platform safety.

📃 Grok's AI-driven IT Operations is now accessible through Microsoft Azure Marketplace. This provides Azure users with an early warning system capable of predicting infrastructure issues before they become full-blown problems and impact customers, assuring smoother app development and the shaping of business strategies.

We love to hear back from you!