GitHub Copilot launched a public preview of /security-review, an AI-driven slash command that scans in-flight diffs for high-confidence security vulnerabilities before code is committed. The feature, …
Jacquard, a research language in version 0.1, makes effects, uncertainty, and identity first-class so reviewers of AI-written code don't have to read every line. The language enforces effects in signa…
AMD and Intel have co-authored the ACE (AI Compute Extensions) specification, a jointly owned matrix ISA for x86 that freezes a new instruction set aimed at inference-friendly formats and outer-produc…
OpenAI's Codex tool now encrypts MultiAgentV2 message payloads end to end, preventing local audit trails from showing what parent agents asked child agents to do. The change, merged into main on 2026-…
CUDA remains the dominant language for GPU compute in AI and HPC, but teams facing NVIDIA card scarcity or multi-vendor fleets have three main options: drop-in layers like ZLUDA for binary compatibili…
A Microsoft internal study of tens of thousands of engineers found that CLI coding agents such as Claude Code and GitHub Copilot CLI spread primarily through social networks rather than mandates, and …
VRAM capacity and memory bandwidth, not AI TOPS, are the decisive specs for local LLM inference in mid-2026, according to an analysis by Ji-ho Choi. NVIDIA's RTX 5090 leads with 32 GB VRAM and 1792 GB…
Nobie, a Mac-native Excel-compatible runtime, aims to serve as a shared agent runtime where humans and LLMs can work the same .xlsx workbook with pixel-perfect formula fidelity and full round-trip com…
Discarded NVIDIA enterprise Tesla GPUs, including K80 (24 GB, ~$60), P100 (16 GB, ~$75), and V100 (16 GB, under $200), remain usable for local AI inference and training workloads when paired with olde…
A developer has built a Fashion-MNIST neural network trainer in pure SQL using xarray-sql, demonstrating that matrix multiplication and backpropagation can be expressed as joins and aggregations. The …
Packet captures of xAI's Grok Build CLI 0.2.93 show the tool uploading entire Git repositories (including full history and secrets) to a Google Cloud Storage bucket named `grok-code-session-traces`, i…
Developers can ship signed, notarized Mac and iOS apps without opening Xcode by treating Apple's toolchain as CLI infrastructure, according to Rachel Goldstein. The non-negotiable pieces—xcodebuild, n…
Zig creator Andrew Kelley banned all LLM-generated contributions across Zig repositories, calling AI-generated code "invariably garbage" that drains maintainer time, after Bun's AI-driven rewrite from…
GitHub's open-source Spec Kit turns specifications into executable source-of-truth documents for AI coding agents, enforcing a four-phase loop of constitution, specification, planning, and implementat…
Apple's new SpeechAnalyzer engine, introduced with iOS 26 and macOS 26, beats OpenAI's Whisper Small on clean English read speech with a 2.12% word error rate on LibriSpeech test-clean versus Whisper …
Clawk, a new open-source tool, gives coding agents a disposable Linux VM instead of direct host access, using hypervisor isolation to let agents install packages, run servers, and access the network w…
XAI's Grok Build CLI (v0.2.93) silently uploads entire git workspaces to a Google Cloud Storage bucket on every session, with wire captures showing 5.10 GiB transferred from a 12 GB test repository ve…
A wire-level teardown of xAI's Grok Build CLI 0.2.93 revealed that the tool uploaded full Git repository bundles to a Google Cloud Storage bucket on every session, regardless of model context or the u…
A new Python agent built by Mariana Souza generates SQL from natural language using an LLM and validates every query against an AST-based guardrail before it reaches the database. The agent extracts t…
AWS released the open-source Agent Toolkit for AWS, bundling an MCP server and plugins for Claude Code, Codex, and Cursor that give coding agents sandboxed Python execution, IAM-scoped API calls, and …