NEWS IN BRIEF: AI/ML FRESH UPDATES

Get your daily dose of global tech news and stay ahead in the industry! Read more about AI trends and breakthroughs from around the world

Revolutionizing Investment Research with AI Assistants

AI-powered assistants using Generative AI and large language models (LLMs) are transforming investment research in capital markets by automating tasks and boosting productivity for financial analysts. These assistants can gather and analyze multifaceted financial data from various sources, enabling analysts to work faster and more accurately in making investment decisions.

Sora's AI Commercial Sparks Controversy at Toys “R” Us

Toys "R" Us partners with Native Foreign to create brand film using OpenAI's Sora text-to-video tool, showcasing founder's story. Sora generates AI video clips of realistic scenes and characters from text instructions, pushing boundaries of technology in commercial production.

AI Al Michaels: Your Daily Olympics Recap

NBC plans to use AI-generated clone of Al Michaels' voice for daily Olympic recaps, sparking concerns about future sports commentators. Michaels, known for iconic calls since 1971, will narrate 2024 Summer Olympics on Peacock, despite being dropped from NFL coverage in 2023.

Efficient Numeric Data Classification with C#

Article presents Nearest Centroid Classification for Numeric Data in Microsoft Visual Studio Magazine. Nearest centroid classification is easy, interpretable, but less powerful than other techniques, achieving high accuracy in predicting penguin species.

Living Human Skin Cells Create Smiling Robot

Researchers from University of Tokyo and Harvard create lifelike robotic skin with human cells, capable of conveying emotions. The study explores benefits of using living tissue for robot covering, emphasizing potential for self-repair.

Optimizing Performance: Load Balancer Essentials

Optimal workload distribution in microservice applications is crucial for handling thousands of requests per second. Load balancers and health checks play key roles in distributing requests and maintaining system efficiency.

Ruthless Gang's Crypto Theft Spree

24-year-old Remy Ra St. Felix led a violent gang targeting cryptocurrency savings, committing armed robberies, assaults, and even a kidnapping. Despite only stealing $150,000 in crypto, the gang inflicted harm on 11 victims across four US states, highlighting the dangers of physical crypto theft.

Enhancing LLMs for Self-Driving with LangProp

ChatGPT powers autonomous driving research at Wayve using LangProp framework for code optimization without fine-tuning neural networks. LangProp presented at ICLR workshop showcases LLM's potential to enhance driving through code generation and improvement.

Unlocking the Future of Chip Designs with 3D Visualization

Ansys utilizes NVIDIA technology to tackle complex challenges in 3D-IC design at the Design Automation Conference. Using NVIDIA Omniverse and Modulus, Ansys engineers can optimize chip performance and reliability with AI-based surrogate models for faster simulations.

Revolutionizing AI: Matrix-Free LLMs

Researchers from UC Santa Cruz, UC Davis, LuxiTech, and Soochow University have developed an AI language model without matrix multiplication, potentially reducing environmental impact and operational costs of AI systems. Nvidia's dominance in data center GPUs, used in AI systems like ChatGPT and Google Gemini, may be challenged by this new approach using custom-programmed FPGA chips.

Guardrails for Responsible AI

Large language models (LLMs) enable human-like conversations but can also spread misinformation and harmful content. Guardrails are crucial for mitigating risks in LLM applications, ensuring safe and desirable outputs.