The AI landscape continues to evolve rapidly, with new investments, innovative model releases, and expert analyses shaping the future of technology. Here are the key developments from today.
Lightspeed Targets $250M for New India Fund, Focusing on Early-Stage AI
Lightspeed Venture Partners is seeking to raise $250 million for a new fund dedicated to early-stage AI investments in India. This marks the first time the Silicon Valley firm is synchronizing its fundraising efforts in India with its global funds, reflecting a strategic shift towards a shorter investment cycle.
Fastino Releases GLiNER2.5-Decide: A 340M Open-Weight Decision Model
Fastino Labs has unveiled GLiNER2.5-Decide, a decision-making model with 340 million parameters that processes text and structured questions to provide accurate, structured answers. Each response includes a probability distribution and a confidence score, aimed at enhancing decision-making in agent pipelines such as routing and triage.
Black Forest Labs Releases FLUX 3 Action: A 7B Open-Weights World Action Model
Black Forest Labs has launched FLUX 3 Action, a 7 billion parameter open-weights model designed for robotics. This model analyzes camera feeds and robot states to predict future actions, achieving record performance on the RoboLab-120 benchmark and running significantly faster than its predecessors.
Top AI Experts Underestimate the Field’s Rapid Advancement
A recent study by the Forecasting Research Institute reveals that leading AI experts have consistently misjudged the pace of AI development. Notable milestones, such as AI achieving gold-medal status at the International Mathematical Olympiad, occurred much earlier than anticipated, though predictions for practical applications like self-driving cars remain mixed.
BottleCap AI Releases ThinkingCap-Qwen3.8-27B: Improved Efficiency
BottleCap AI has introduced ThinkingCap-Qwen3.8-27B, a fine-tuned model that reduces thinking token usage by 37.2% while maintaining high accuracy. This model offers a drop-in replacement for existing frameworks and boasts improvements in long-context performance, making it a valuable tool for developers.
Sakana AI Hires Jürgen Schmidhuber as Chief Scientific Advisor
Tokyo-based Sakana AI has appointed deep learning pioneer Jürgen Schmidhuber as its Chief Scientific Advisor. Known for his groundbreaking work in AI, Schmidhuber will lead the company’s new RSI Lab, focusing on recursive self-improvement, a concept that allows AI systems to enhance their own capabilities over time.
Google’s Suncatcher Project Explores Solar-Powered AI Data Centers in Orbit
Google is launching the Suncatcher project, which aims to establish AI data centers in orbit powered by solar energy. An experimental satellite is set to launch soon, but experts note the ambitious project would require thousands of satellites to match the energy output of a single terrestrial data center.
Meta’s Muse Charm Blends AI with Fashion Trends
Meta has introduced Muse Charm, an AI gadget that resembles a Tamagotchi while tapping into Gen Z’s fascination with retro tech and fashion accessories. This innovative device aims to engage users by combining playful design with advanced AI functionalities.
Google Photos Introduces AI-Powered Virtual Closet Feature
Google Photos has rolled out a new AI feature that creates a virtual wardrobe from users’ photos, inspired by the film ‘Clueless.’ This functionality is now available on both Android and iOS, allowing users to organize and visualize their clothing options effortlessly.
Compiled automatically by the Tech AI Newsdesk from public AI-news sources and summarised in our own words.
