Full-text search across 69272 articles. Combine with topic and date filters; results sorted by relevance.
A developer built a full-stack document translator entirely through AI-driven 'Vibe Coding,' converting legal and bureaucratic texts into everyday Portuguese while enforcing WCAG AAA accessibility and LGPD compliance. Th…
More than 100 organizations, including OpenAI, Anthropic, AWS, Google, Microsoft, Oracle, Cloudflare, CrowdStrike, and Hugging Face, have signed an open letter warning that AI-enabled cyberattacks will become 'far more w…
Terence Tao uses large language models not as mathematical oracles but as tools to manage information density, summarizing technical sections and writing verification code, according to a practical guide for high-level r…
StackGen's State of Reliability Report, analyzing 109,000+ incidents, found that unregulated AI agents caused at least nine cases of data destruction over the past year, and median resolution times have remained flat sin…
Researchers introduced Procedura, a 3D modeling agent framework that uses an LLM to write objects as parametric, procedural assemblies with named parts and machine-checkable mates. In evaluations on P3D-Bench and the new…
MonkeyCode, an open-source AI coding project offering free models and a free server, warns that shared AI servers have a concurrency ceiling that can be measured in 30 minutes before teams hit latency collapse. The artic…
OpenAI announced on August 28, 2026, that it will wind down the contract supplying models to the AI coding tool Cursor, highlighting that most AI products are resellers licensing models from labs like OpenAI, Anthropic, …
XPress, a lightweight causal refiner for diffusion drafters in speculative decoding, raises acceptance length by ~30% on average (up to +56%) and decoding throughput by ~1.3× on average (up to 1.7×) compared to the dFlas…
Chinese AI company Z.ai has confirmed that the anonymous 'Ox Alpha' model that topped OpenRouter's popularity charts last week is its new GLM-5.3-Flash, an open-weight model now available across official platforms and on…
A tutorial from Quickchat AI demonstrates building an AI scheduling assistant for Calendly that answers questions, books, reschedules, and cancels meetings in one conversation, using Calendly's hosted MCP server and Quic…
Google has launched Gemini 3.7 Flash, a successor to Gemini 3.6 Flash, designed to improve instruction following, intent clarification, and multi-step planning for coding and agent-based tasks. The model is available glo…
Abhishek launched Opslane, an open-source agent that combines error tracking and session recording to identify user-facing issues and automatically create pull requests only when it can verify a fix. The tool, built afte…
JetBrains Research has open-sourced DPTrainer, a library that integrates Opacus with Hugging Face's Trainer to enable differential privacy (DP-SGD) training without rewriting training loops or modifying trainer source co…
Xebia, a global AI-first digital transformation and engineering partner with 5,000 professionals across 16 countries, is hiring a Senior Full-stack Developer (AI-Native) for an on-site role in Bulgaria, Poland, or Romani…
Google announced two new Android app quality requirements this week, including performance thresholds to reduce memory usage and code optimization, in response to industry-wide memory chip shortages driven by the AI data…
The Linux Foundation has submitted its Open Model, Data, and Weights license (OpenMDW-1.1) to the Open Source Initiative for review and potential approval as an AI open-source license, aiming to address the lack of a bro…
Knosh 0.3.0, a one-shot coding agent built on the Koog framework, implements a permission system that allows agents to deny specific tools by their unique identifiers in frontmatter, with filesystem-manipulating tools re…
Google will not automatically penalize AI-generated content, as the company evaluates pages based on substance and user behavior signals rather than origin, according to an analysis by Dejan SEO. The article argues that …
Alibaba released a preview of Qwen3.8-Flash-Next, a 125-billion-parameter model that activates only 6 billion parameters per token, achieving training costs roughly one-ninth of typical models of its scale. The model rep…
Four papers published today challenge the reliability of LLM-as-judge, showing that judge annotations can mark working code as failures, leak across rubric dimensions, and misalign with human creativity judgments, while …