Veo 2 is now available to Gemini Advanced users with enhanced realism, prompt-based video generation, mobile uploads to TikTok and YouTube, and image-to-video animation via Whisk Animate.
DolphinGemma, an AI model co-developed by Google, Georgia Tech, and WDP, decodes dolphin vocalizations using 40 years of acoustic data. It supports real-time interaction and open scientific research.
Hugging Face has acquired Pollen Robotics to launch an open-source robotics platform, integrating Reachy 2 and LeRobot to enable AI-powered machines for all.
ByteDance’s Seaweed-7B achieves real-time AI video with audio-video sync, long-form storytelling, and 2K upscaling — all with 7B parameters and low GPU needs.
Nvidia will manufacture AI chips and supercomputers in the U.S., investing $500B through 2029. New factories, robotics, and Omniverse tech power the move.
OpenAI launched GPT-4.1, mini, and nano with 1M-token input limits and strong coding capabilities. These models rival Google’s Gemini and Anthropic’s Claude.
Netflix is trialing an OpenAI-integrated search feature in Australia and New Zealand, enabling users to discover content through personalized, natural language queries.
At Google Cloud Next 2025, Google unveiled Ironwood TPU, expanded Gemini across platforms, boosted Vertex AI by 20x, introduced Unified Security, and reinforced its AI-first cloud strategy.
OpenAI has enhanced ChatGPT with a persistent memory feature, allowing the AI to remember and utilize information from previous user conversations to provide more personalized and context-aware responses.
