Google launches Veo on Vertex AI for text/image to video creation and makes Imagen 3 globally available for high-quality image generation. Both models feature SynthID watermarking, enhancing security and enterprise AI workflows.

AWS re:Invent unveiled Amazon's Nova AI models, including Nova Micro, Lite, Pro, and the upcoming Premier in 2025. New image and video generators with watermarking were introduced. Amazon is partnering with Anthropic to build the largest AI cluster and plans to release voice models and a revamped Alexa next year.

ByteDance’s Doubao updates include image comprehension with new photo and camera features on mobile and PC, enabling accurate image recognition and queries on landmarks and movies. It also launched video generation. Doubao excels in market influence but seeks to enhance user experience amid higher expectations.

Cohere launches Rerank 3.5, an advanced AI search model supporting over 100 languages. It enhances reasoning, integrates seamlessly with enterprise systems, boosts search accuracy by up to 30%, and excels in cross-language queries, benefiting industries like finance and technology.

Tencent launches Hunyuan, a 13B-parameter open-source AI video model, the largest to date. Available on Yuanyou APP and Tencent Cloud APIs, it features hyper-realistic visuals, high semantic consistency, and smooth motion. Fully open on GitHub and Hugging Face, it supports developers with comprehensive prompts and advanced technology for superior video creation.

Intel CEO Pat Gelsinger retires after over 40 years. David Zinsner and Michelle Holthaus appointed as interim co-CEOs while Intel searches for a successor. The company faces AI competition and financial pressures from IDM2.0 but secured $7.86B under the CHIPS Act.

Liquid AI's STAR framework uses evolutionary algorithms to optimize AI architectures, surpassing Transformers in efficiency and performance for language models by creating smaller, high-performing models.

World Labs, founded by Fei-Fei Li, developed an AI system that creates interactive 3D environments from a single image, allowing keyboard and mouse navigation. Backed by $230M and valued over $1B, the startup aims to launch its first product in 2025.

AI chip startup Tenstorrent raises $693M in Series D, valuing it at $2B. Aiming to challenge Nvidia with cost-effective AI hardware based on Tensix core and RISC-V architecture, it attracts major investors. Funds will support open-source AI stack and global expansion.

The Browser Company CEO Josh Miller announced Dia, an AI-focused browser launching early 2025. Dia integrates AI deeply into browsing with features like content suggestions, idea generation, and task automation. Support for the existing Arc browser will continue.

OpenAI CFO Sara Frier revealed the company is considering an ad-based revenue model but will proceed cautiously to ensure ads don't impact user experience. Currently, there are no concrete plans, as OpenAI prioritizes advancing AI and delivering high-quality services.

ChatGPT recently failed to recognize "David Mayer," triggering online speculation about potential censorship or data removal. Users explored theories like blacklisting or privacy laws, while OpenAI investigates the anomaly.

AWS unveils AI enhancements for Amazon Connect, featuring customer segmentation, automated marketing campaigns, integration with Amazon Lex and Q, Salesforce and WhatsApp support, and AI-driven performance evaluation. These upgrades enhance support efficiency for thousands of organizations managing over 10 million daily interactions.

Apple released iOS 18.1.1 and introduced "Image Playground" in iOS 18.2, an AI tool that transforms text into cartoon-style images. Accessible via Apple Intelligence and Messages, it allows customization through text or personal photos. Available in iOS 18.2 beta, with a December rollout expected.

Elon Musk's legal team seeks a preliminary injunction against OpenAI’s shift to for-profit, alleging antitrust violations with Microsoft, such as restricting investments and misusing sensitive data. OpenAI denies the claims as unfounded.

WeChat Official Accounts backend launches "AI Image" feature, enabling creators to generate images from text descriptions. It provides four automatic options, customizable styles and aspect ratios, enhances efficiency, and reduces copyright risks by producing unique visuals.

Tsinghua University's AIR is beta testing China’s first AI Hospital, Agent Hospital, launching by year-end and opening to the public next year. It features 42 AI doctors across 21 departments handling over 300 diseases, aiming for efficient, high-quality care. AI doctors rapidly develop through simulations for real-world integration. Other AI hospitals include Shanghai Weiyi and Eastern Hospital’s AI models.

Virginia University's SMAST, an AI-driven video analyzer, accurately detects human behaviors in real time. This technology enhances surveillance, public safety, healthcare motion tracking, and autonomous vehicle navigation, setting new standards in action detection.

Anthropic secures $8B from AWS, enhances AI infrastructure with Trainium, launches enterprise AI platform Claude on Amazon Bedrock, introduces high-performance Claude 3.5 models, and deploys on GitHub Copilot, advancing responsible AI innovation.

Meta’s Chief AI Scientist Yann LeCun predicts AGI within a decade, emphasizing the need for novel architectures beyond current models. Despite LLM limitations, he remains optimistic about superintelligence breakthroughs by 10 years.