Chain of News Digest

Chain of News 18/11/2026

18/11/2026
**Top Story** Nvidia’s interim CEO announced that the chipmaker will cease all sales of its GPUs to mainland China, effectively ending a decade‑long market that accounted for roughly 15% of its revenue. The decision follows the abrupt departure of founder Jensen Huang, whose leadership had kept the China pipeline alive despite U.S. export restrictions. By pulling out, Nvidia signals a hardening of the tech‑cold war, forcing Chinese AI startups to scramble for alternative hardware or accelerate domestic chip programs. For developers, the move means the loss of a familiar, high‑performance platform for training large models, pushing them toward AMD, Intel, or emerging Chinese silicon solutions. The shift also reshapes the global supply chain, as Nvidia redirects capacity to Europe and the U.S., potentially tightening availability for existing customers elsewhere. SOURCES: [1] **AI Models & Research** The “Split-and-Scale Latent 3D Representations” paper introduces a hierarchical latent space that can encode entire 3D scenes in a fraction of the memory required by voxel grids, enabling real‑time generative editing of complex environments. Developers building immersive AR/VR or simulation pipelines will now have a more tractable way to integrate generative 3D content without prohibitive storage costs. Another noteworthy advance is the open‑source AI strategy reaffirmed by Nvidia’s interim leadership, which includes a planned acquisition of Hugging Face to cement a unified model hub and tooling ecosystem. While not a research paper, this strategic move will accelerate the diffusion of cutting‑edge models and lower integration friction for developers seeking ready‑to‑use transformers. SOURCES: [7], [2] **Developer Tools & Frameworks** Vapi, Retell AI, and LiveKit each offer distinct approaches to building voice agents: Vapi and Retell AI provide fully managed orchestration with low‑latency APIs, while LiveKit’s open‑source Agents framework hands developers deeper control over signaling and media pipelines. This comparative analysis helps engineers choose between turnkey services for rapid prototyping and a customizable stack for bespoke applications. UiPath’s new AI Coding Agent blends Claude Code and Codex, allowing developers to describe desired code snippets in natural language and receive production‑ready implementations, dramatically shortening the iteration loop for routine automation scripts. Meanwhile, OpenAI’s ChatGPT for Sales plugin suite adds over 30 partner integrations that let developers embed prospect research, pitch generation, and CRM updates directly into conversational workflows, opening a fast path to AI‑augmented sales tools without building backend data pipelines. SOURCES: [4], [6], [5] **Industry & Business** Amazon and Google are set to collectively own nearly 30% of Anthropic after its upcoming IPO, with Amazon holding just under 15% and Google close behind at 14%, positioning the two cloud giants as dominant shareholders in a leading LLM provider. This stake not only secures preferential access to Anthropic’s models for their respective cloud platforms but also intensifies the rivalry for AI talent and enterprise contracts. In parallel, Nvidia’s interim leadership reiterated that the company’s open‑source AI roadmap remains unchanged despite Huang’s exit, signaling continuity for developers who rely on Nvidia’s SDKs, libraries, and the forthcoming Hugging Face integration to streamline model deployment across heterogeneous hardware. SOURCES: [3], [2] **Worth Watching** The rapid consolidation of AI talent among cloud providers, exemplified by Amazon and Google’s Anthropic stakes, may soon force smaller startups to seek niche differentiators or align with alternative clouds, reshaping the competitive landscape for AI SaaS. Nvidia’s withdrawal from China could accelerate the emergence of a parallel AI hardware ecosystem in Asia, prompting developers to maintain dual‑stack codebases to stay portable across divergent chip architectures. Finally, the emergence of AI coding assistants like UiPath’s agent hints at a broader trend toward “code‑by‑conversation” interfaces, which could redefine how development teams allocate junior versus senior resources in the near term. SOURCES: [3], [1], [6]

Today's Stories

Today's articles

TechCrunch AI

Nvidia is abandoning China after Jensen Huang’s ouster

Nvidia's interim CEO says the company is done selling chips to China. How much will that cost?

17/11/2026
TechCrunch AI

Nvidia says its open source AI strategy remains intact after Jensen Huang’s exit

Nvidia's interim CEO reaffirmed its open source AI strategy and plans to acquire Hugging Face, despite Jensen Huang's sudden departure.

17/11/2026
TechCrunch AI

Amazon and Google will together own nearly 30% of Anthropic after its IPO

Amazon will own just under 15% of Anthropic after the IPO, with the option to go much higher. Google is a close second with holdings of nearly 14%.

17/11/2026
DZone

Vapi vs Retell AI vs LiveKit: Architecture, Latency, Pricing, and Developer Experience

Vapi, Retell AI, and LiveKit can all power voice agents, but they put different quantities of infrastructure and application logic in the developer's hands. Vapi and Retell AI are managed orchestration platforms. LiveKit Agents is an open-source, code-first framework that can run on LiveKit Cloud or infrastructure you operate. That distinction affects customization, deployment, integration design and total cost, not just which dashboard you prefer.

17/11/2026
OpenAI Blog

Introducing ChatGPT for Sales

Explore the new Sales plugin and 30+ new and expanded partner plugins in ChatGPT to research prospects, prepare pitches, update your CRM, and move deals forward.

17/11/2026
GNews: AI Agents Code

UiPath Launches AI Coding Agent, Integrates Claude Code and Codex to Target Development Automation - BigGo Finance

UiPath Launches AI Coding Agent, Integrates Claude Code and Codex to Target Development Automation BigGo Finance

16/11/2026
Apple ML Research

Split-and-Scale Latent 3D Representations

Latent 3D representations have significantly expanded the capabilities of modern shape modeling, enabling compact encoding of 3D scenes and supporting powerful 3D generative models. Despite this progress, current representations struggle to capture the fine details of complex objects, fundamentally limited by the capacity of their latents.

16/11/2026