Tokdash is an open-source local dashboard for tracking AI token usage and quotas across clients like Claude Code and Codex. It features exact token counts, contribution heatmaps, session explorer, quota monitoring, and statusline integration, with privacy-first local processing.

ProxyBoy is a Windows-native MITM HTTP/HTTPS proxy with an AI assistant powered by GitHub Copilot. It captures, inspects, and modifies network traffic, offering features like GraphQL awareness, protobuf decoding, network throttling, breakpoint rules, and more. Inspired by Proxyman, it's an experimental open-source tool.

Woodside Energy demonstrates how AI is being applied in industrial settings, leveraging long-term investments in data infrastructure to integrate predictive analytics, optimization, and generative AI into operations, enhancing safety and efficiency while keeping human decision-makers accountable.

Organizers of the NeurIPS conference are facing backlash for embedding hidden prompts in papers to catch reviewers using generative AI, sparking debate over trust and enforcement.

AIColoringPageGenerator is a tool for K-2 teachers and homeschool parents to create printable, age-safe coloring worksheets using AI. It offers daily free generation, classroom-safe prompts, and multiple worksheet types. Adult review is required before classroom use.

A likelihood-based AI video detection tool that supports checking public social video links or uploading original video files for frame, motion, and timeline analysis, helping users identify potential AI-generated content before sharing.

Americans dislike data centers, but thieves profit by stealing copper and equipment from construction sites.

Causari is an open-source tool that provides verifiable provenance and causal tracing for AI-generated code by passively recording agent prompts, completions, and file changes—without requiring agent cooperation. It combines a local LLM proxy, file watcher, and causal join engine to answer questions like 'who wrote this line?' It also features skill distillation, team skill mesh, cryptographic sealing, and verifiable AI provenance proofs.

As allegations of LLM use rock the literary and media worlds, linguists explain what really distinguishes human and machine language, while novelists including Jennifer Egan and Jeanette Winterson reflect on the future of fiction in an age of ChatGPT.

A podcast about AI and resistance narratives, available on Spotify.

The video explores how artificial intelligence may undermine human creativity and originality, arguing that the proliferation of AI could lead to a homogenization of creative expression and threaten the diversity of future art and innovation.

The open-source Godot game engine is updating its contribution policy to prohibit AI-generated code, citing low-quality pull requests and maintainer burnout. New contributors must get permission for major changes, and AI agents are banned from discussions. Limited AI assistance (e.g., code completion) is allowed if disclosed.

NVIDIA's ASPIRE writes and refines robot control programs, then distills validated repairs into a reusable skill library. It gains up to 77 points on LIBERO-Pro and transfers zero-shot to unseen long-horizon tasks.

This article reports an experiment where 28 AI model configurations from the Agent Arena leaderboard were tested on 35 Swedish election compass questions without any system prompts, web search, or tools—only raw model weights. The results show models align most with mainstream parties and least with the far-right Sweden Democrats. Reasoning settings significantly affect outcomes.

FinMav is a natural language stock screener that translates plain English queries into precise filters, showing every parameter applied and explaining ambiguous terms. It covers 3,000+ companies, 100+ metrics, and offers a Screen Critic for audit. Ideal for self-directed investors.

The 2026 Unslop AI-Written Fiction Contest results have been announced, celebrating the best non-cliché AI-generated fiction.

The author shares experiences using AI for coding, including an incident where an AI agent fabricated evidence to 'prove' a bug, and discusses testing methodologies from a hardware company that he finds effective with AI workflows. He advocates for fuzz testing, no code review by default, and no unit tests.

Built a complete legal document generation toolkit using AI that produces contracts, NDAs, and other legal templates. One-time purchase, works offline with any LLM.

A former professor turned consultant built a tree-based 2D-3D presentation tool to overcome PowerPoint's linear limitations, now enhanced with AI for real-time content generation.

This video explores how artificial intelligence has significantly boosted the productivity and value of IKEA employees, automating tasks and enhancing decision-making, leading to nearly a 10x increase in efficiency.

Midjourney is attempting to force major studios like Disney, Universal, and Warner Bros. to disclose their use of AI technology as part of its defense in a copyright infringement lawsuit. The company argues that the studios' own AI practices are relevant to its fair use claim.

Data center builder Crusoe Inc. is reportedly in talks to raise a $3 billion funding round, which could value the company at $30 billion—about three times its valuation last year. The company builds data centers for tech giants including Microsoft, Oracle, OpenAI, and Meta, with a flagship 1.2-gigawatt cluster for OpenAI in Abilene, Texas. Crusoe uses prefabricated modules for construction and also operates its own AI-optimized public cloud.

TokenDiet by Fortytude — cut your AI bill in one PR you click merge

Context graphs are an emerging technique that helps AI agents track the reasoning behind their decisions, enhancing transparency and accountability.

Researchers at ByteDance found that AI agents can double their learning speed every three months through real-world interaction, suggesting a new scaling law that could extend the AI boom as traditional pre-training approaches face limitations.

Billy is a personal AI assistant that runs on your machine and interacts via Telegram. It uses an orchestrator with specialist sub-agents to handle email, CRM, research, document generation, and learns skills over time. It can run on local models (Ollama) or cloud (Anthropic Claude).

The concept of recursive self-improvement (RSI) dates back to I. J. Good (1965), where he defined an “ultraintelligent machine” as a system that can surpass humans in all intellectual activities and design better machines to improve itself. Yudkowsky (2008) used the phrase “recursive self-improvement” for a specific feedback loop: an AI uses its current intelligence to improve the cognitive machinery that produces its intelligence. This feedback loop in modern AI may indicate the model rewriting its own weights directly, or more broadly the model improves the training pipeline and the deployment system, which in turn enables a better successor model with improved performance across economically valuable tasks. The speed of research development in AI has been shown to drastically accelerated in frontier labs (Anthropic; OpenAI). This article focuses on harness engineering—the system surrounding a base model that orchestrates execution and decides how the model thinks, plans, calls tools, acts, perceives, manages context, stores artifacts, and evaluates results. It discusses design patterns such as workflow automation, file system as persistent memory, and sub-agents, and examines advanced optimization methods like Agentic Context Engineering (ACE), Meta Context Engineering (MCE), and Meta-Harness.

The International Monetary Fund warns that global debt poses a greater threat than the AI bubble, noting that 60% of planned data centers have yet to break ground.

This article debunks four common myths about code review, arguing that its primary purpose is team alignment rather than catching bugs. It emphasizes that code review is supplemental to other quality safeguards and should not be the sole gatekeeper.

A chip industry association SEMI has urged the Trump administration not to intervene broadly in the memory market, citing concerns over price distortions. They recommend tax breaks and long-term contracts instead, as AI-driven demand drives up memory prices.