Alibaba Cloud has released the latest entry in its growing Qwen family of large language models. The new Qwen with Questions (QwQ) is an open-source competitor to OpenAI’s o1 reasoning model. As with competing large reasoning models (LRMs), QwQ can correct its own mistakes, relying on extra compute cycles during inference to assess its responses, making it well suited for reasoning tasks like math and coding. Described as an “experimental research model,” this preview version of QwQ has 32-billion-parameters and a 32,000-token context, leading to speculation that a more powerful iteration is in the offing. Continue reading Qwen with Questions: Alibaba Previews New Reasoning Model

Hailuo, the free text-to-video generator released last month by the Alibaba-backed company MiniMax, has delivered its promised image-to-video feature. Founded by AI researcher Yan Junjie, the Shanghai-based MiniMax also has backing from Tencent. The model earned high marks for what has been called “ultra realistic” video, and MiniMax says the new image-to-video feature will improve output across the board as a result of “text-and-image joint instruction following,” which means Hailuo now “seamlessly integrates both text and image command inputs, enhancing your visuals while precisely adhering to your prompts.” Continue reading MiniMax’s Hailuo AI Rolls Out New Image-to-Video Capability

Alibaba Cloud last week globally released more than 100 new open-source variants of its large language foundation model, Qwen 2.5, to the global open-source community. The company has also revamped its proprietary offering as a full-stack AI-computing infrastructure across cloud products, networking and data center architecture, all aimed at supporting the growing demands of AI computing. Alibaba Cloud’s significant contribution was revealed at the Apsara Conference, the annual flagship event held by the cloud division of China’s e-retail giant, often referred to as the Chinese Amazon. Continue reading Alibaba Cloud Ups Its AI Game with 100 Open-Source Models

Backed by Alibaba and Tencent, Chinese startup MiniMax has launched a new text-to-video model called Hailuo AI that is quickly gaining traction on social media based on its impressive capabilities, with comments ranging from “fantastical” to “hyper-realistic.” The free, web-based tool has already triggered videos that have gone viral, despite the current limitation of only 6-second clips. However, an image-to-video model is reportedly coming soon, in addition to a version 2 that promises longer video duration and improved motion. Unlike the Jimeng AI text-to-video model that was issued by ByteDance last month, the MiniMax technology is available outside of China. Continue reading Hailuo AI: China’s MiniMax Releases Free Text-to-Video App

China’s largest cloud computing company, Alibaba Cloud, has released a new computer vision model, Qwen2-VL, which the company says improves on its predecessor in visual understanding, including video comprehension and text-to-image processing in languages including English, Japanese, French, Spanish, Chinese and others. The company says it can analyze videos of more than 20 minutes in length and is able to respond appropriately to questions about content. Third-party benchmark tests compare Qwen2-VL favorably to leading competitors and the company is releasing two open-source versions with a larger private model to come. Continue reading Alibaba’s Latest Vision Model Has Advanced Video Capability

Google DeepMind has unveiled new research on AI tech it calls V2A (“video-to-audio”) that can generate soundtracks for videos. The initiative complements the wave of AI video generators from companies ranging from biggies like OpenAI and Alibaba to startups such as Luma and Runway, all of which require a separate app to add sound. V2A technology “makes synchronized audiovisual generation possible” by combining video pixels with natural language text prompts “to generate rich soundscapes for the on-screen action,” DeepMind writes, explaining that it can “create shots with a dramatic score, realistic sound effects or dialogue.” Continue reading DeepMind’s V2A Generates Music, Sound Effects, Dialogue

Nvidia is expanding its substantive influence in the AI sphere with Nemotron-4 340B, a family of open models designed to generate synthetic LLM training data for commercial applications across numerous fields. Through what Nvidia is calling a “uniquely permissive” free open model license, Nemotron-4 340B provides a scalable way for developers to build LLMs. Synthetic data is artificially generated data designed to mimic the characteristics and structure of data found in the real world. The offering is being called “groundbreaking” and an important step toward the democratization of artificial intelligence. Continue reading Nvidia’s Open Models to Provide Free Training Data for LLMs

Alibaba is touting a new artificial intelligence system that can animate portraits, making people sing and talk in realistic fashion. Researchers at the Alibaba Group’s Institute for Intelligent Computing developed the generative video framework, calling it EMO, short for Emote Portrait Alive. Input a single reference image along with “vocal audio,” as in talking or singing, and “our method can generate vocal avatar videos with expressive facial expressions and various head poses,” the researchers say, adding that EMO can generate videos of any duration, “depending on the length of video input.” Continue reading Alibaba’s EMO Can Generate Performance Video from Images

As China’s tech giants begin releasing ChatGPT-style products, the powerful Cyberspace Administration of China (CAC) has issued a draft of what will become the country’s first set of restrictions for generative AI. Ground rules include limiting content created by AI to that which reflects the core values of socialism and does not subvert state power in text or image, according to the draft. Models should also be trained in a way that does not allow them to generate false information or exhibit bias as to race, ethnicity or gender. The news comes as three Chinese firms release generative AI platforms. Continue reading Chinese Government Drafts New Rules for Generative AI Tech

SoftBank-owned Alibaba Group — with headquarters in Hangzhou, China — plans to split into six independent companies that may seek separate IPOs, the company announced as Chinese authorities appear to be winding down a regulatory clampdown on the country’s powerful technology sector. The business empire assembled around e-commerce by industrialist Jack Ma these past 20 years was at its peak valued at more than $800 billion but is now assessed at about one-fourth that amount. The company’s stock soared on the news adding about $32 billion in global value, a rising tide that also lifted competitors’ boats. Continue reading Alibaba to Split into Six New Companies with Potential IPOs

Instagram will cease allowing product tags for live-stream shopping in the U.S. as of March 16, a functionality that has been available to creators and businesses since 2020. Although live-stream shopping is popular and profitable in Asian markets, it has been slow to take hold in the U.S. and Europe. Instagram says users will still be able to set up shops and leverage shopping opportunities across their feeds, stories and Reels, just not in live broadcasts. The company will “continue to invest in shopping experiences,” focusing on those “that provide the most value to our users.” Continue reading Instagram Will End Live-Stream Shopping and Focus on Ads

Live-stream shopping has been slow to take off in the U.S., but some feel the sector is at a tipping point. U.S. revenue from the format is projected to reach $20 billion this year and grow to $57 billion in 2025 says Coresight Research. However, that’s a drop in the bucket compared to the activity in Asia. McKinsey Digital says outlets like Alibaba can rack up more than $7 billion in 30 minutes. Estimates place the Asia-Pacific market at about $180 billion in live-streamed shopping for 2021. Now big players like Walmart, Target, YouTube and TikTok are getting serious about cracking open the U.S. market, too. Continue reading Retailers, Big Tech Get Behind Live-Stream Shopping in U.S.

Senate Homeland Security Committee leaders Gary Peters (D-Michigan) and Rob Portman (R-Ohio) have introduced a bill requiring a risk framework for open-source code. The proposed legislation would require the Cybersecurity and Infrastructure Security Agency to develop the risk evaluation process for open-source software being used by federal agencies and critical infrastructure. The move follows the discovery in December of a vulnerability in the Apache Software Foundation’s popular Log4j Java logging utility. Peters said the Log4j incident presented a serious threat to banks, hospitals, and utility companies, among other national security operations. Continue reading Senate Group Wants CISA to Protect Open-Source Software

China’s tech sector is taking a hit. Revenue for e-commerce giant Alibaba posted what appears to be its most sluggish quarterly increase ever, at 9 percent, with competitor JD.com also paced among its worst performances, with revenue up 18 percent for Q1. The nation’s search engine leader, Baidu, reported 1 percent revenue growth, while Tencent remained flat. Supply chain problems due to the COVID-19 resurgence in conjunction with Beijing’s recent measures to battle back monopolies are cited as causes for the negative market turn. Since late 2020, China has been investigating alleged monopolistic practices among many top corporations. Continue reading Chinese Technology Companies Are Experiencing Slowdowns

NBCUniversal is expanding its partnership with social video platform TikTok to include a joint promotion for the 2022 Olympic Winter Games in Beijing, set to kick off on February 4, and the 2022 Paralympic Winter Games, beginning March 4. NBC Winter Games advertisers will reportedly have first crack at piggybacking onto what is being couched as a new type of brand experience that lets them draw from NBC 2022 Winter Games clips as they try to connect with sports fans. TikTok videos with Olympic-related hashtags have resulted in 18 billion views on the platform thus far, according to NBC. Continue reading NBCUniversal Teams with TikTok to Promote Winter Olympics