🧠 OpenAI co-founder Andrej Karpathy joins Anthropic 🧠
The legendary AI researcher who co-founded OpenAI in 2015 and led Tesla's Autopilot vision team brings his expertise to Anthropic at a pivotal moment for frontier labs.

Andrej Karpathy, one of the most recognized figures in modern AI research, has officially joined Anthropic. A co-founder of OpenAI in 2015, he later led Tesla's Autopilot vision team from 2017 until 2022, briefly returned to OpenAI, then left in 2024 to launch Eureka Labs, an AI-native education startup, while becoming one of the most followed AI educators online through his "Neural Networks: Zero to Hero" lectures. Now he joins the pre-training team under Nick Joseph and will lead a new internal effort to apply Claude to Anthropic's own training pipeline. Citing the next few years at the LLM frontier as "especially formative," his move represents a major competitive pressure shift among leading labs and strengthens Anthropic's research bench at a critical moment as the race for frontier capabilities intensifies.

🤖 Gemini 3.5 Flash ships for agentic workflows 🤖
Google's newest model targets agentic speed and enterprise-wide rollout.

Google has officially released Gemini 3.5 Flash, announced at Google I/O 2026 as its next-generation model built for agentic workflows, coding, and long-horizon task execution. Designed for efficiency and multi-step task completion, Flash serves as a workhorse across both consumer and enterprise applications — applying direct competitive pressure on OpenAI and Anthropic's frontier models. Google is expanding Gemini 3.5 Flash access across its entire ecosystem, including Search, Android Studio, enterprise tools, and developer platforms. This deep integration signals a strategy of embedding agentic AI at every touchpoint. It's one of the most significant model releases of 2026 so far, setting a new bar for what lightweight frontier models can accomplish.

💰 OpenAI sells guaranteed API compute to enterprises 💰
Enterprise customers can lock in dedicated compute with multi-year pricing commitments.

OpenAI has introduced a new Guaranteed Capacity offering, enabling enterprise customers to secure long-term access to dedicated compute for powering AI products, agents, and workflows. Customers can choose between one-, two-, and three-year commitments, with deeper discounts for longer terms. The offering will remain available until OpenAI's current allocation is sold out. This mirrors cloud infrastructure pricing models pioneered by AWS and Azure, marking a strategic shift toward recurring revenue and infrastructure-as-a-service — a strong adoption signal for enterprise AI. The move strengthens OpenAI's enterprise positioning and builds a competitive moat through long-term customer lock-in.

INTERESTING TO KNOW

🎬 Gemini Omni generates video from any input 🎬

At Google I/O 2026, Google announced Gemini Omni, a truly multimodal model whose rollout enables transforming text, images, audio, or video inputs directly into video outputs. This capability jump moves beyond single-modality generation into cross-modal video synthesis, positioning Google at the forefront of an increasingly contested space. The model could reshape content creation, education, and marketing workflows as video generation becomes a core frontier for leading AI labs.

🎥 NVIDIA opens LongLive 1.0 for real-time video 🎥

NVIDIA has released LongLive 1.0, an open framework for interactive long-form video generation with real-time user-guided editing and sequential prompting. This rollout addresses a key limitation of current video models: producing coherent, lengthy content with user control. Available on GitHub, the framework leverages streaming attention and KV-cache optimization, positioning NVIDIA as an active software contributor beyond its hardware business — a competitive pressure move that could accelerate adoption for creative professionals and game developers.