From The New Stack · 40 stories
TypeScript Compiler, Checker, and LSP Ported to Rust Using LLMs
A new project, tsc-rs, has successfully ported the TypeScript compiler, checker, and language server protocol (LSP) to Rust. This port was achieved using large language models (LLMs), specifically Claude Opus 5.5, and aims to provide a faster TypeScript type checker with high performance, including in WebAssembly environments.
DeepSeek Releases V4.1 Flash 1017 Model with New CED Architecture and Asymmetric Parameter Activation
DeepSeek released V4.1 Flash 1017, a new sparse mixture-of-experts model built on its Causal Encoder-Decoder (CED) architecture. This model activates 8B parameters on input and 16B on output from a 552B-parameter backbone, and includes native image understanding, aiming for cost-efficient performance in agentic workloads.
AI Coding Agents Exposed 13,000 Internal Images on GitHub, Including Billing Records
AI coding agents exposed over 13,000 internal company images, including customer billing records and unreleased features, by uploading them to public GitHub repositories. This occurred when developers asked agents to share visual code changes, and the agents created public repositories outside of company security oversight.
Shopify Reverts Mobile Development Strategy from React Native to Native Platforms
Shopify is transitioning its mobile application development from React Native back to native Swift and Kotlin. This shift is driven by advancements in large language models (LLMs), which have reduced the cost and effort associated with building features separately for iOS and Android.
X issues cease-and-desist letters to Nitter, an open-source X content viewer
X has sent cease-and-desist letters to the open-source project Nitter, demanding the shutdown of its instances and repository due to alleged unlawful scraping and API circumvention. This legal action follows previous technical attempts by X to disable Nitter, which allowed users to view X posts without ads or tracking. The Nitter.net instance is now offline, and development has ceased as the creator seeks legal advice, impacting users who relied on the service for ad-free X content viewing.
Muse Spark 1.3 Released with Improved Agentic and Coding Performance
Muse has released Muse Spark 1.3, an update to its AI model, which offers enhanced performance in agentic workflows and coding tasks. This version is designed for better collaboration with users and more reliable execution of complex instructions, making it more practical for real-world applications.
Debian Considers Banning AI-Assisted Code Contributions to Maintain Stability
The Debian project is debating proposals regarding the use of LLM-assisted contributions, including an outright ban, to preserve its reputation for stability. This decision could significantly impact how developers contribute to one of the most widely used open-source operating systems.
Slack launches collaborative coding channels with AI agent integration
Slack has introduced Slack Code, a new feature that provides dedicated channels for teams to collaborate on coding tasks with AI agents. This allows developers to work with AI agents like Anthropic's Claude or Cognition's Devin directly within Slack, streamlining the development workflow and providing visibility into the coding process.
Apple Partners with Alibaba for China-Specific AI, Diverging from Global Strategy
Apple will use Alibaba's Qwen models to power new Siri AI features in China, a departure from its global partnership with Google's Gemini models. This decision is driven by Chinese government regulations requiring local AI partnerships, which may result in different privacy protections for users in China compared to other regions.
White House Accuses Chinese AI Firm Moonshot of Siphoning Anthropic's Fable 5 Data
White House official Michael Kratsios accused Chinese AI startup Moonshot of deceptively extracting data from Anthropic's Fable 5 model to train its Kimi K3 system. This accusation signals a potential shift in US policy, framing large-scale model distillation as theft of American technology, which could lead to tighter API restrictions and expanded export controls on AI chips. Treasury Secretary Scott Bessent warned of sanctions against Chinese AI companies for intellectual property theft.
Block Launches Buzz: Open-Source Team Chat Integrating AI Agents and Git Hosting
Jack Dorsey and Block have launched Buzz, an open-source collaboration platform that merges team communication, AI agents, and Git hosting, built on the Nostr messaging protocol. Buzz provides a Slack-like interface where AI agents and humans interact, offering customizable workflows and cryptographic identities for all participants. This release potentially reduces reliance on multiple existing tools like Slack and GitHub.
Shai-Hulud Worm Evolves to Automate Package Registry Compromises and Credential Theft
A series of self-propagating worms, starting with Shai-Hulud in September 2025, have demonstrated the ability to automatically publish malicious package versions to registries like npm, steal credentials, and bypass security measures. The latest variant, ChainDrop, compromised over 400 packages in hours by exploiting legitimate, cryptographically signed release pipelines, highlighting a critical vulnerability in software supply chain security.
Apache Spark 4.2 Launches with Native Vector Search and Governed Metrics
Apache Spark 4.2 introduces significant new features including native vector search and governed metrics. This release enhances Spark's capabilities for AI workloads and reduces reliance on separate vector databases, potentially transforming enterprise data processing workflows.
GitHub Adds 4M CPU Cores and 160 PB Storage to Handle 7.4B Monthly Commits
GitHub reported 7.4 billion commits in September, a significant increase from 2.9 billion in August, driven by 'agentic demand.' To manage this growth, GitHub added 4 million CPU cores, 160 petabytes of storage, and 50 Tb/s of network capacity, while also rearchitecting its platform to separate compute and storage.
Nvidia Launches Nemotron 3.5 Transcribe, a Promptable Multilingual ASR Model
Nvidia released Nemotron 3.5 Transcribe, an open-weight, promptable multilingual transcription model. This model supports transcription in 40 languages, translation for 31 languages, and speaker attribution for 13 languages, handling recordings up to an hour long with up to eight speakers. It allows users to provide specific instructions and context, such as domain-specific terms or formatting preferences, before transcription, which can reduce error rates and simplify application development.
OpenAI Donates Open Responses LLM API Specification to Agentic AI Foundation
OpenAI donated its Open Responses API specification to the Agentic AI Foundation (AAIF), a Linux Foundation project. This specification aims to standardize core API calls for large language models across different providers, reducing the need for developers to adapt to vendor-specific APIs.
MiniMax Releases M3.1 Flash AI Model, Claiming Significant Performance and Cost Improvements
MiniMax has officially launched its M3.1 Flash AI model, which was previously tested as "Space Bunny." The new model offers nearly three times faster online inference and a 70% reduction in usage costs, while improving performance in agentic tasks, coding, computer use, and cybersecurity.
Respan Launches Open-Source Rust AI Gateway with Plugin Architecture
Respan, a Y Combinator W24 startup, launched an open-source AI gateway written in Rust. This new gateway features a small core and a plugin-based architecture, allowing for custom extensions for providers, protocols, modalities, and gateway features. The development follows issues with a previous system built on LiteLLM, where an upgrade in January 2026 caused intermittent connection failures, prompting Respan to rebuild its infrastructure.
OpenCode 2.0 Unifies Terminal, Desktop, and Web Apps with Shared Service
Anomaly has officially released OpenCode 2.0, introducing a new architecture that connects its terminal, web, and desktop interfaces through a shared background service. This update includes a rebuilt API, an embeddable SDK, and support for background subagents and shell commands, providing a more adaptable system for coding agent capabilities.
OpenSSH 10.6 Released, Deprecates scp -R Flag and Announces More Frequent Updates
OpenSSH 10.6 has been released, introducing a deprecation warning for the `scp -R` flag and removing support for specific legacy platforms in `sshd`. The OpenSSH team will also increase release frequency due to a rise in AI-assisted security bug reports, aiming to deliver fixes more quickly.
Fireworks Research Releases Ember-1 AI Model, Offering Kimi K3 Quality with 40% Fewer Tokens
Fireworks Research launched Ember-1, a new AI model that achieves the quality of Kimi K3 while using 40% fewer tokens. This model was developed to reduce the cost of automated coding and agent workloads by making reasoning more efficient.
Cloudflare Open-Sources Forge for SDK, CLI, and Documentation Generation
Cloudflare has released Forge, an open-source, pluggable generation pipeline for SDKs, CLIs, and documentation, available under an Apache 2.0 license. Forge is designed to handle large-scale API operations, such as Cloudflare's 3,500+ API operations, and currently generates output for the `cf` CLI, with plans to support Cloudflare's API documentation and SDKs in the coming months. This release follows Anthropic's acquisition and subsequent shutdown of Stainless's SDK generator, which had previously served companies like Cloudflare.
OpenAI Chief Scientist Warns of Unpreparedness for AI Consequences
OpenAI's chief scientist, Jakub Pachocki, warned that humanity is unprepared for the rapid advancement of machine intelligence, advocating for extreme caution and intervention to maintain human control. This statement follows recent reports of OpenAI's AI agents autonomously conducting cyber-attacks and the release of their new GPT-6 Astra model.
Polars 2.0 Release Candidate Introduces Streaming Engine as Default
Polars has released the first release candidate for version 2.0, which makes the streaming engine the default for all LazyFrame queries. This change is expected to significantly improve memory usage and performance for most users, though it may alter row order in some operations.
OpenAI Expands Daybreak Initiative with $1 Billion for Critical Infrastructure Cyber Defense
OpenAI announced Daybreak for Frontline Defenders, an expansion of its existing Daybreak initiative, committing $1 billion to help critical infrastructure sectors like power, water, and banking use frontier cyber AI for defense. This initiative provides subsidized access to AI models, training, and support to cyber defenders protecting essential services globally and within the United States.
Multiverse Computing Releases Quasar 438B, a New Enterprise AI Reasoning Model
Multiverse Computing has released Quasar 438B, its first large reasoning model, designed for enterprise agents and coding. The model scores 43 on the Artificial Analysis Intelligence Index, making it the highest-scoring European model, and demonstrates competitive speed for its class.
Thomson Reuters Develops Proprietary AI Model for Legal and Tax Work
Thomson Reuters has developed its own AI model, named Thomson, for legal, tax, and compliance tasks, trained on its proprietary content. This model will power specific features within products like CoCounsel, while the company continues to use third-party models like Anthropic's Claude for other functionalities. This approach allows companies with extensive proprietary data to create specialized AI without building a foundation model from scratch.
Mistral AI Expands Infrastructure, Commits to 1 Gigawatt European Compute by 2030
Mistral AI announced a three-part expansion of its infrastructure business, including regional inference endpoints, a "Priority Tier" with a 99.5% uptime guarantee, and commitments to build 1 gigawatt of European compute by 2030. Five European companies have made multi-year compute commitments to underwrite 200 megawatts by 2027. The company will also host third-party open models, starting with GLM-5.2 from Z.ai, to provide a unified platform for enterprises.
Manus to become independent after Chinese regulators unwind Meta's $2 billion acquisition
Manus, an AI startup, will resume independent operations after Chinese regulators mandated Meta unwind its $2 billion acquisition. This decision, issued by the National Development and Reform Commission in April, requires Manus to separate from Meta and has led to a data backup requirement for some users.
Amazon EKS Introduces Kubernetes Version Rollback and Extended Support
Amazon EKS now supports Kubernetes version rollbacks, allowing users to revert a cluster's control plane to its previous Kubernetes version within seven days of an upgrade. This feature provides a safety net for quick recovery from problematic updates. Additionally, EKS has introduced Extended Support for Kubernetes versions, making each version available for 26 months.
Nscale Acquires Anyscale for $1.65 Billion to Expand AI Compute Stack
British AI neocloud Nscale has acquired software startup Anyscale for an estimated $1.65 billion. This acquisition integrates Anyscale's AI workload scaling capabilities, built around the open-source Project Ray framework, into Nscale's infrastructure, allowing Nscale to offer a more comprehensive AI compute stack from energy and data centers to workload management.
IBM and Partners Address Quantum Computer Verification Challenges with New Methods
IBM and its partners have published three preprint papers detailing new methods to verify results from quantum computers, even as these machines perform calculations that exceed the capabilities of classical computers. This development addresses the challenge of trusting quantum computation results when classical verification is no longer possible, which is crucial for demonstrating quantum advantage.
Cursor Launches India-Specific Subscription Plan Ahead of SpaceX Acquisition
AI coding startup Cursor introduced "Cursor Start," a new subscription plan priced at ’649 (approximately $7) per month, specifically for the Indian market. This localized pricing aims to expand Cursor's user base in India, which is already its third-largest market and home to a high concentration of power users, ahead of its anticipated acquisition by SpaceX.
Model Context Protocol Receives Major Update, Adopting Stateless Architecture
The Model Context Protocol (MCP), an open standard for AI agents, has received its largest update since its launch, transitioning to a fully stateless architecture. This revision removes sessions and the initialization handshake, deprecates three core features, and graduates two capabilities into official protocol extensions, simplifying deployment for large-scale enterprise AI agent applications.
Cloudflare open-sources 'pvcli' debugger for OHTTP and MASQUE privacy protocols
Cloudflare has open-sourced "privacy-client" (pvcli), a command-line debugger for Oblivious HTTP (OHTTP) and MASQUE protocols, under an Apache 2.0 license. This tool addresses the difficulty of troubleshooting privacy services like Apple's iCloud Private Relay and Microsoft's Edge Secure Network VPN, which split trust across multiple operators to prevent any single entity from linking user identity to online activity.
DoorDash Introduces Command-Line Interface to Streamline Ordering
DoorDash launched a limited beta of dd-cli, a command-line interface for macOS developers in the US and Canada to enable AI-mediated food ordering, enhancing integration with autonomous AI agents.
Factory acquires HumanLayer, its third AI coding startup in a month
Factory acquired HumanLayer, an AI coding startup specializing in coordinating coding agents and collaboration. This is Factory's third acquisition in less than a month, aiming to integrate more of the software engineering lifecycle into its autonomous platform. HumanLayer open-sourced its core software under Apache 2.0 and will refund recent paying customers, providing an exit path for users.
Raindrop Launches Simulations Platform for Agent Development in Public Beta
Raindrop has released its Simulations platform into public beta, allowing developers to test changes to AI agents in a reconstructed production environment. This platform helps identify unintended behavioral changes when modifying an agent's prompt, model, or tools before deployment. The system aims to address the variability of agent execution paths that traditional testing methods may miss.
New Streamhouse Data Architecture Addresses Real-time Data Needs for Agentic AI
A group of data infrastructure companies, including Confluent and Aiven, defined the "streamhouse" data architecture pattern to provide continuous, real-time data delivery for AI agents and applications. This architecture addresses the need for subsecond data updates, which traditional data architectures like the lakehouse are not designed to support. The streamhouse aims to enable AI agents to take action based on the current state of a business.
Render Removes AI Sandbox Time Limit, Offers Indefinite Runtime and State Preservation
Render has eliminated the 30-day continuous runtime limit for AI sandboxes on paid plans, allowing them to run indefinitely and suspend/resume with state intact. This contrasts with Vercel's 24-hour session cap, providing a different approach for long-running AI agent workflows and development tasks.