Alibaba is making AI news again, releasing another Qwen reasoning model, QwQ-32B, which was trained and scaled using reinforcement learning (RL). The Qwen team says it “has the potential to enhance model performance beyond conventional pretraining and post-training methods.” QwQ-32B, a 32 billion parameter model, “achieves performance comparable to DeepSeek-R1, which boasts 671 billion parameters (with 37 billion activated),” Alibaba claims. While parameters refer to the total set of adjustable weights and biases in the model’s neural network, “activated” parameters are a subset used for a specific inference task, like generating a response. Continue reading Alibaba Says Qwen Reasoning Model on Par with DeepSeek

Sesame, an AI startup from Oculus co-founder Brendan Iribe, has created a conversational voice model that many feel has achieved uncanny levels of authenticity. Drawing comparisons to the charismatic vocal centerpiece of the 2013 Warner Bros. film “Her,” Sesame seems to have achieved a new level of engagement among AI voice assistants. While some are describing the tech as “amazing.” others have expressed concern over its capabilities. “Our goal is to achieve ‘voice presence’ — the magical quality that makes spoken interactions feel real, understood and valued,” explains a blog post by Iribe and others. Continue reading AI Startup Sesame Develops Next Stage of Voice Generation

Alibaba has open-sourced its Wan 2.1 video- and image-generating AI models, heating up an already competitive space. The Wan 2.1 family, which has four models, is said to produce “highly realistic” images and videos from text and images. The company has since December been previewing a new reasoning model, QwQ-Max, indicating it will be open-sourced when fully released. The move comes after another Chinese AI company, DeepSeek, released its R1 reasoning model for free download and use, triggering demand for more open-source artificial intelligence. Continue reading Highly Realistic Alibaba GenVid Models Are Available for Free

Barely two weeks after the launch of its OmniHuman-1 AI model, ByteDance has released Goku, a new artificial intelligence designed to create photorealistic video featuring humanoid actors. Goku uses text prompts to create among other things, realistic product videos without the need for human actors. This last is a boon for ByteDance social media unit TikTok. Goku is open source, trained on a large dataset of roughly 36 million video-text pairs and 160 million image-text pairs. Goku’s debut is received as more bad news for OpenAI in the form of added competition, but a positive step for global enterprise. Continue reading ByteDance’s Goku Video Model Is Latest in Chinese AI Streak

OpenAI has decided to simplify its product offerings. A month after announcing the in-development GPT-o3 as its next frontier model, the company has canceled it as a standalone release, explaining that it would be integrated into the upcoming GPT-5 instead. “A top goal for us is to unify o-series models and GPT-series models by creating systems that can use all our tools, know when to think for a long time or not, and generally be useful for a very wide range of tasks,” OpenAI co-founder and CEO Sam Altman wrote in a social media post this week. Expected to ship later this year, the GPT-5 models will incorporate voice, canvas, search, deep research and more, OpenAI says. Continue reading Sam Altman Reveals Plans to Simplify OpenAI’s Product Line

Anthropic is launching an AI Economic Index aimed at understanding the technology’s effects on labor markets and the economy over time. The Anthropic Economic Index kicks off with a report that indicates the most concentrated uses of AI today are in software development and technical writing tasks. The information was culled by analyzing data from millions of anonymized conversations on Claude.ai with the goal of revealing how AI is being incorporated into real-world tasks across global markets. It also found that AI usage trends more toward augmenting human capabilities (57 percent), compared to using it for automation (43 percent). Continue reading Anthropic Launches Economic Index to Track the AI Economy

Meta is seeking to make AI more inclusive with a program to support underserved languages “and help bring their speakers into the digital conversation.” Meta’s Fundamental AI Research (FAIR) unit has teamed with UNESCO to launch the Language Technology Partner Program, which is looking for people who can provide more than 10 hours of speech recordings (with transcriptions) and chunks of written text (200+ sentences, with translation) in diverse languages. “Partners will work with our teams to help integrate these languages into AI-driven speech recognition and machine translation models, which when released will be open sourced,” Meta said. Continue reading Meta Adds Indigenous Languages to Speech and Translation AI

Most people know Hugging Face as a resource-sharing community, but it also builds open-source applications and tools for machine learning. Its recent release of vision-language models small enough to run on smartphones while outperforming competitors that rely on massive data centers is being hailed as “a remarkable breakthrough in AI.”  The new models — SmolVLM-256M and SmolVLM-500M — are optimized for “constrained devices” with less than around 1GB of RAM, making them ideal for mobile devices including laptops and also convenient for those interested in processing large amounts of data cheaply and with a low-energy footprint. Continue reading Hugging Face Has Developed Tiny Yet Powerful Vision Models

Jack Dorsey’s financial tech and media firm Block (formerly Square) has released a platform for building AI agents: Codename Goose. Previously available in beta, Goose is primarily designed to build agents for coding and software development, but Block built in many basic features that could be applied to general purpose pursuits. Because it is open source and offered under Apache License 2.0, the hope is that developers will apply it to varied use cases. A leading feature of Codename Goose is its flexibility. It can integrate a wide range of large language models, letting developers use it with their preferred model. Continue reading Codename Goose: Block Unveils Open-Source AI Agent Builder

Meta is rolling out personalization updates to its Meta AI personal assistant. At the end of last year, the company introduced a feature that lets Meta AI remember what you’ve shared with it in one-on-one chats on WhatsApp and Messenger so it could produce more relevant responses. That feature will now be available to Meta AI on Facebook, Messenger and WhatsApp for iOS and Android in the U.S. and Canada. “Meta AI will only remember certain things you tell it in 1:1 conversations (not group chats), and you can delete its memories at any time,” explains the company. Continue reading Facebook, Instagram, WhatsApp Get Meta AI Memory Boost

Hangzhou-based AI firm DeepSeek is roiling the U.S. tech sector and upending financial markets. The startup has managed to become competitive with Silicon Valley’s deep learning firms despite U.S. sanctions that prevent Chinese technology companies from buying premium chips. DeepSeek has made it into the global top 10 in terms of model performance, and as of this week had the top-ranked free AI assistant at the Apple App Store. DeepSeek’s new R1 model has drawn attention for using less computing power than competing systems, while performing comparably, despite having been developed using older Nvidia chips. Continue reading Chinese AI Startup DeepSeek Disrupting the U.S. Tech Sector

Nvidia Cosmos, a platform of generative world foundation models (WFMs) and related tools to advance the development of physical AI systems like autonomous vehicles and robots, was introduced at CES 2025. Cosmos WFMs are designed to provide developers a way to generate massive amounts of photo-real, physics-based synthetic data to train and evaluate their existing models. The goal is to reduce costs by streamlining real-world testing with a ready data pipeline. Developers can also build custom models by fine-tuning Cosmos WFMs. Cosmos integrates Nvidia Omniverse, a physics simulation tool used for entertainment world-building. Continue reading CES: Nvidia’s Cosmos Models Teach AI About Physical World

Samsung Electronics has teamed with Google on a new spatial sound standard, Eclipsa Audio, that could emerge as a free alternative to Dolby Atmos. On display at CES 2025 in Las Vegas this week, the format is rolling out across Samsung’s line of 2025 TVs and soundbars, and Google will support it on the content side by enabling Eclipsa 3D audio on some YouTube videos this year. Samsung has been a notable holdout on Dolby Vision HDR embracing instead the competing HDR10+. Now the South Korean electronics giant seems to be staking out its own turf in 3D audio, advocating for open source. Continue reading CES: Samsung and Google Team on Spatial Audio Standard

Meta’s FAIR (Fundamental AI Research) team has unveiled recent work in areas ranging from transparency and safety to agents, and architectures for machine learning. The projects include Meta Motivo, a foundation model for controlling the behavior of virtual embodied agents, and Video Seal, an open-source model for video watermarking. All were developed in the unit’s pursuit of advanced machine intelligence, helping “models to learn new information more effectively and scale beyond current limits.” Meta announced it is sharing the new FAIR research, code, models and datasets so the research community can build upon its work. Continue reading Meta Rolls Out Watermarking, Behavioral and Concept Models

Meta Platforms has packed more artificial intelligence into a smaller package with Llama 3.3, which the company released last week. The open-source large language model (LLM) “improves core performance at a significantly lower cost, making it even more accessible to the entire open-source community,” Meta VP of Generative AI Ahmad Al-Dahle wrote on X social. The 70 billion parameter text-only Llama 3.3 is said to perform on par with the 405 billion parameter model that was part of Meta’s Llama 3.1 release in July, with less computing power required, significantly lowering its operational costs. Continue reading Meta’s Llama 3.3 Delivers More Processing for Less Compute