Trending AI news stories + papers

Today's Top AI news stories and papers

AI news stories

1. US, Britain, other countries ink agreement to make AI 'secure by design'

   Several countries, including the United States and Britain, have introduced a non-binding international agreement to prioritize the security of artificial intelligence (AI) technology. The 20-page document outlines the importance of designing and deploying AI systems with a focus on public safety, offering general recommendations like monitoring AI for misuse and protecting data integrity. While not legally binding, this agreement represents a significant step in addressing AI safety concerns globally. It includes 18 countries and aims to prevent AI from being exploited by malicious actors. Europe is also working on AI regulations, and the United States is pushing for AI regulation amid concerns about its impact on various aspects of society.

2. Nvidia Delays Rollout Of China-Focused AI Chip

   Nvidia is delaying the release of its new AI chip, the H20, in order to comply with U.S. export restrictions on sales to China. The chip's launch has been postponed to the first quarter of 2024. This news caused Nvidia's stock to drop by 1.9% in morning trade. U.S. trade restrictions on China have been a growing concern for Nvidia investors, even though the company recently reported strong quarterly results. Nvidia's CFO, Colette Kress, warned that these restrictions will negatively impact their China business, and the long-term magnitude of the impact is uncertain.

3. Wikipedia founder Jimmy Wales says AI is a ‘mess’ now but can become superhuman in 50 year

   Jimmy Wales, the founder of Wikipedia, expressed his concerns about the limitations of ChatGPT-4, stating it's not suitable for writing Wikipedia articles due to its errors and fabrications. He believes that achieving superhuman AI capabilities could take 50 years or more. Wales emphasized the importance of governance in startups, even in the early stages. While critical of current AI, Wales hasn't ruled out the possibility of using AI to improve Wikipedia's accuracy. He's open to partnerships with open-source AI companies but stressed the need for caution.

4. AI can figure out sewing patterns from a single photo of clothing

   Researchers at Sea AI Lab in Singapore have developed an artificial intelligence model that can generate accurate sewing patterns for garments from a single photo of the clothing being worn. Sewing patterns are typically used by clothing makers as templates to cut and sew fabric pieces to create a garment, and recreating these patterns from existing clothing can be a labor-intensive process. The AI model is designed to streamline this process and make it more efficient.

5. Meet the first Spanish AI model earning up to €10,000 per month

   Aitana, an AI-generated virtual influencer created by The Clueless agency, has emerged as a novel trend in influencer marketing. A response to challenges faced by traditional models and influencers, Aitana offers brands a customizable, cost-effective, and controversy-free representation. While this AI-driven approach disrupts the modeling industry and reduces costs, it also raises ethical concerns about promoting unrealistic beauty standards. Nonetheless, it reflects the evolving landscape of influencer marketing and its potential to redefine industry dynamics.

AI Repos

1. huggingface / transformers

   Transformers provides thousands of pretrained models to perform tasks on different modalities such as text, vision, and audio

2. eosphoros-ai / DB-GPT

   DB-GPT is an open-source framework designed for the realm of large language models (LLMs) within the database field. Its primary purpose is to provide infrastructure that simplifies and streamlines the development of database-related applications.

3. NVIDIA / Stable-Diffusion-WebUI-TensorRT

   This extension enables the best performance on NVIDIA RTX GPUs for Stable Diffusion with TensorRT.