Reinforcement learning hits a wall with reasoning models because every token gets the same reward. A new algorithm from Alibaba’s Qwen team fixes this by weighting each step based on how much it shapes what comes next, doubling the length of thought processes in the process.
Netflix has open-sourced an AI framework that can remove objects from videos and automatically adjust the physical effects those objects had on the rest of the scene. The system is called VOID, short for "Video Object and Interaction Deletion." What makes it special is that beyond erasing objects from a scene, it also handles the downstream physical effects, like collisions, that the removed object originally caused.
VOID is built on top of Alibaba's CogVideoX video diffusion model, fine-tuned with synthetic data from Google's Kubric and Adobe's HUMOTO for interaction detection. Google's Gemini 3 Pro analyzes the scene and identifies affected areas, while Meta's SAM2 handles segmenting the objects that need to be removed. An optional second pass uses optical flow to correct any shape distortions.
The project was developed by Netflix researchers in collaboration with INSAIT Sofia University. Code, paper, and demo are available on GitHub, arXiv, and Hugging Face. The system ships under the Apache 2.0 license, which means it can be used commercially.
Several leadership changes are underway at OpenAI. Fidji Simo, CEO of the newly created "AGI Deployment" division, is taking sick leave for several weeks to deal with an autoimmune disease affecting her nervous system. While she's away, OpenAI President Greg Brockman will take over product responsibilities, including the company's super app plans. On the business side, CSO Jason Kwon, CFO Sarah Friar, and CRO Denise Dresser will step in.
Head of Marketing Kate Rouch is also stepping down for health reasons. Rouch plans to return in a smaller role once her health improves. Gary Briggs will fill in as her temporary replacement.
COO Brad Lightcap is stepping down as well, moving to a new "special projects" team reporting directly to CEO Sam Altman. Dresser is picking up most of his responsibilities. Lightcap's work on government relations and "OpenAI for Countries" shifts to the strategy department.
Anthropic has acquired AI biotech startup Coefficient Bio for roughly 400 million dollars in shares, according to Newcomer and The Information. The startup was founded just eight months ago, had fewer than ten employees, and built a platform using AI for pharmaceutical tasks like drug research planning and identifying new drug opportunities.
The real draw is likely the talent. Anthropic appears to be after specialists who can accelerate its pharma work immediately: the team is joining Anthropic's Healthcare and Life Sciences division under Eric Kauderer-Abrams. Venture capital firm Dimension, which held about half of the startup, is reporting a 38,513 percent return. Both companies declined to comment.
The acquisition comes as big pharma ramps up its own AI efforts. Eli Lilly recently signed a billion-dollar deal with AI drug developer Insilico Medicine. Anthropic already works with Sanofi, Novo Nordisk, and AbbVie, while Google DeepMind has spun off its own AI medicine company.
Deepseek v4 is expected to launch in the coming weeks, and it will run entirely on Huawei chips. According to The Information, the model represents a major milestone in China's effort to break free from foreign chip dependency. Deepseek reportedly spent months working with Huawei and chip designer Cambricon to port the model to Chinese-made chips. Notably, Nvidia didn't get early access to v4—only Chinese chip companies did.
The bet on domestic hardware might already be paying off. Chinese tech companies including Alibaba, Bytedance, and Tencent have ordered hundreds of thousands of units of Huawei's new Ascend 950PR to run Deepseek v4 through their cloud services and integrate it into their own AI applications, according to five people familiar with the matter. The surge in demand pushed chip prices up by 20 percent.
Huawei says the Ascend 950PR delivers roughly 2.8 times the computing power of Nvidia's H20, though it still falls short of the H200. Huawei also continues to face production bottlenecks caused by US export controls.
Microsoft is investing $10 billion in Japan from 2026 to 2029, its largest ever commitment to the country. Vice Chairman and President Brad Smith presented the plans during a visit to Tokyo. The funding covers AI infrastructure, cybersecurity, and workforce training.
Together with SoftBank and Sakura Internet, Microsoft plans to offer GPU-based AI services through Azure, with all data staying in Japan. This should make it possible to develop Japanese language models. On the security side, Microsoft is deepening its partnership with Japan's cybersecurity agency and the National Police to better detect and prevent cyberattacks.
Japan is projected to be short 3.26 million AI and robotics specialists by 2040, according to the Ministry of Economy. To address that, Microsoft is teaming up with Fujitsu, Hitachi, NEC, NTT Data, and SoftBank to train one million engineers and developers by 2030. Microsoft had already invested $2.9 billion in Japan in 2024.
Anthropic has looked into complaints from users who were hitting their Claude Code usage limits much faster than expected. According to Anthropic's Lydia Hallie, tighter limits during peak hours and sessions with 1-million-token contexts growing larger are the two main reasons for the problem. Hallie says Anthropic also fixed some bugs, but none of them led to incorrect billing. The company has also shipped efficiency improvements and added in-product pop-ups to keep users informed.
Hallie recommends using Sonnet 4.6 instead of Opus, since Opus burns through limits roughly twice as fast. She also suggests turning off Extended Thinking when it's not needed, starting fresh sessions instead of continuing old ones, and limiting the context window. Users who still notice unusually high usage should report it through the feedback function.
