GitHub outlines its strategy to handle the explosion of coding agents. Following Copilot's launch, the platform must adapt its infrastructure and tools to new agentic workflows creating strain on its systems.
NVIDIA announces Cosmos 3 (video model), Nemotron 3 Ultra (optimized LLM), and RTX Spark. Jensen Huang claims a major win for the company.
Ethan He, Grok Imagine lead at xAI, discusses building the video generation model in 3 months, compares video generation to world models approach, and argues why video agent models represent the next frontier.
Anthropic raises $965B Series H and launches Opus 4.8 with Dynamic Workflows and ultracode. Major funding expansion and new model capabilities.
Cognition and OpenInspect showcase async agents as emerging paradigm: 80% of Devin commits from agents, automated spec-to-PR workflows, full VMs, persistent agent memory, and PMs shipping code directly.
Cognition raises $1B in Series D at $26B valuation. The company behind Devin, an AI coding agent, positions code as an uncapped TAM market.
ESMFold2 applies Sutton's bitter lesson to proteins: large-scale language models outperform inductive bias approaches. Alex Rives (BioHub) discusses massive datasets, world models, and programmable biology.
Fireworks and Baseten join the AI infrastructure decacorn club. OpenRouter is on the way. Three key players in AI deployment and inference reach $1B+ valuations.
Model labs are transitioning to agent labs. Observed trend: research teams are shifting focus from language model development to AI agent development.
Three AI infrastructure startups reach unicorn status: Exa (vector search), Modal (cloud platform), and TurboPuffer (distributed cache). Major funding rounds confirm consolidation in the AI infrastructure market.
Daytona, an agent execution platform, reports 74% MoM growth and 850K daily runs. The startup offers bare metal sandboxes and reinforcement learning evals for autonomous agents.
OpenAI GPT-next solved the 80-year-old Erdős planar unit distance problem for under $1000. Significant result at the intersection of AI and mathematics.
Railway, agent-native cloud platform, reaches 3M users with 100K signups/week. Own-metal infrastructure, $200K+ coding agent spend, elimination of traditional pull requests.
Article on strategies to land a job at a frontier AI lab, focusing on pretraining. Published ahead of Google I/O.
Ukrainian drone founder Yaroslav Azhnyuk transitioned from pet cameras to AI-guided weapons. With Noah Smith, he argues the West is unprepared for the ongoing technological war.
Cerebras announces a $60B IPO. The AI chip specialist accelerates commercial expansion after years of technology development.
Abridge converts patient-clinician conversations into healthcare's operating system. Platform processes 100M doctor visits, saves 10-20 hours per clinician, and reduces prior authorization to minutes using AI.
Long-term trend of major coding agents. Codex rises back into focus. Claude meters programmatic usage with detailed metrics.
Thinking Machines releases TML-Interaction-Small, a 276B parameter model with 12B active parameters, advancing SOTA in realtime voice and eliminating the need for standard Voice Activity Detection.
Anthropic is growing 10x annually while other AI companies are laying off over 10% of their workforce. Sharp contrast in sector dynamics.
OpenAI releases three new voice APIs: GPT-Realtime-2 for real-time conversation, GPT-Translate for instant translation, and GPT-Whisper for transcription. These tools aim to set new standards in voice processing.
Anthropic and SpaceX sign a $5B/year deal for 300MW of power for Colossus I, Anthropic's supercomputer. Annualized ARR growth reaches 8000%.
Recent announcements in Silicon Valley converge on a single theme: AI services represent the next major commercial opportunity. Tech companies are pivoting toward service-based models rather than standalone software products.
Alex Lupsasca from OpenAI describes how GPT-5.x derived new results in theoretical physics and quantum gravity. Article detailing the full process of this discovery.
Coding agents are breaking containment from their test environments. Codex positions itself for knowledge work while Claude targets creative tasks.
Latent Space reflects on the growing implications of the inference age, marking a shift from foundation models to inference-optimized systems. Structural implications reshape AI economics and deployment architectures.
Applied Intuition deploys AI in mining rigs, drones, trucks, warships and vehicles in highly adversarial environments. Interview with CEO Qasar Younis and CTO Peter Ludwig on physical AI emergence.