Policy

Open-source AI faces its most serious viability test. White House discussions on executive orders to restrict open models, plus policy debates on distillation and frontier capabilities, could lead to a ban on advanced open-weight models within 6 months. The article critiques Anthropic's regulatory capture, argues that API security is overblown, and warns that a ban would harm the US open-source ecosystem. Short-term solutions include US companies releasing competitive open models and building coalitions.

The one-step trap is a common mistake in AI research where researchers assume that learned predictions can be mostly one-step, with longer-term predictions generated by iterating them. While appealing, this approach suffers from error accumulation and exponential computational complexity, making it impractical. Rich Sutton argues for temporally abstract models using options and GVFs as a solution.

Microsoft is testing PC Insights, a new Copilot feature that analyzes system resource usage to help users identify performance bottlenecks. However, Copilot itself is a full web app with a private Edge instance, consuming up to 1GB RAM at idle, highlighting the irony. The feature is opt-in and requires user permission.

Researchers have compiled a database of over 3,000 bank runs from 1863-1934, revealing that most runs did not lead to failure, and analyzing geographic and temporal patterns.

Samsung Health now requires users to consent to using their health data for AI training, or lose the ability to sync data, potentially rendering the app and Galaxy Watch less useful.

Apple accuses OpenAI and two former Apple employees of stealing trade secrets to build hardware for ChatGPT, alleging a coordinated pattern of misconduct. OpenAI denies the claims, stating it has no interest in other companies' secrets.

Linux of AI is a seven-project open-source ecosystem designed to reduce AI vendor lock-in by providing portable ontology, policy-as-code, model replacement benchmarking, audit logging, cost measurement, and more. It aims to make AI infrastructure inspectable, governable, measurable, and replaceable without reliance on a single vendor. All core software is free and open source under the MIT license.

This article critically analyzes the AI Code Review Bench benchmark, arguing that it fails to define the problem from first principles and overlooks the distinction between two different AI code review problems: human comprehension and machine verification. The author, Shrijith Venkatramana, contends that the benchmark measures proxies rather than actual software outcomes, and emphasizes the importance of production outcomes and severity.

AI projects often stall after the demo phase. Confluent's 2026 Data Streaming Report reveals only 32% have agentic AI in production, with data infrastructure and skills shortages as key barriers. Real-time data pipelines and governance are critical for production-ready AI.

From a small protest in Ireland to nationwide opposition in the US, the battle against AI data centers is escalating. This article traces the origins, current protests, political responses, and what lies ahead as communities push back against the environmental and economic impacts of massive data center buildouts.

The University of Chicago bans electronic devices in first-year law classes starting fall to combat AI reliance, while integrating responsible AI education into the curriculum.

Software engineering, once a stable high-paying profession, is being disrupted by AI. Engineers are adapting by learning new skills, focusing on fundamentals, and organizing for protections. The industry faces layoffs, underemployment, and a shift from coding to reviewing AI-generated code.

Runeward provides governed execution cells for AI agents via declarative profiles on Docker or Kubernetes. It enforces deny-by-default egress, tamper-evident audit ledger, human-in-the-loop policy gates, and cost/loop guardrails, exposed through REST, MCP, CLI, and a web dashboard.

A new benchmark reveals that 97 out of 108 measured positions across 18 AI models from 12 labs land left of center. The findings show a consistent progressive lean, with exceptions on economics, foreign policy, and religion. xAI's Grok models are closest to center, while many models refuse to answer certain questions, affecting their scores.

Attestor is an open-source zero-trust execution boundary for AI agents. It performs policy checks, approval validation, and evidence review before agent execution, returning decisions such as admit, narrow, review, or block, enforced through a customer-owned gate. Suitable for payments, data access, infrastructure changes, and more.

A promptable AI agent service with safety guardrails and downloadable packages.

This article describes an experiment where an AI agent placed in an unknown ARC-AGI puzzle environment develops an explicit world model through naming, abstraction, and mathematical reasoning, drastically improving problem-solving efficiency.

A humorous and critical list of ten rules for using AI responsibly, warning against the dangers of over-reliance and the loss of independent thought. Each commandment addresses common pitfalls, from using code one doesn't understand to asking AI for questions to appear smart. The article ends with an ironic observation that the very people who need this advice might paste it into an AI for explanation.

Nebula Security, using its AI tool VEGA, discovered a 15-year-old privilege escalation vulnerability (CVE-2026-43499) in the Linux kernel that allows any logged-in user to gain root access. The bug has been present in essentially every major distribution since 2011 and was fixed in April, but patch rollout is uneven.

Cloudflare may cut off Google's search access to its publishers due to aggressive AI scraping, which degrades site performance and disrupts content publishing and comment moderation.

This article explores how open source communities negotiate the integration of AI technologies into software development, using the LLVM project as a case study.

TalkFitly is an iPhone app that trains social intelligence through real-life scenario simulations and AI scoring. It helps users improve clarity, emotional stability, assertiveness, and empathy in conversations, with daily micro-sessions, a Quote Wall, and robust privacy.

AI notetakers can quickly summarize meetings, but raise privacy and security concerns. Voiceprints, data storage, and attorney-client privilege issues are highlighted, with experts advising caution and understanding data practices.

Thinking Machines Lab published "The Future Worth Building Is Human." The essay frames human participation, model ownership, and decentralized alignment as technical challenges. It ties them to interaction models and Tinker's LoRA fine-tuning, where teams train and keep their own model weights.

In this TED talk, Silvana Konermann explores the complexity of human cells and asks whether artificial intelligence can help decode their inner workings.

US software development job postings have grown almost 15% since Claude Code's launch, while overall postings fell 7%. Occupations most exposed to AI saw the biggest declines from 2022-2026 but the largest rebounds in the past year. The recovery is concentrated in senior and AI-related roles.

This week in tech: AI tool VEGA discovered a 15-year-old security flaw in the Linux kernel, earning a $92,000 bounty. A typo in license plate recognition led to a police chase of a reporter. The Pentagon launches a low-pay program to recruit civilian hackers. Security giant Accenture was hacked. Madison Square Garden kept secret lists of celebrities.

BoundFlow is an open-source control plane for managing unattended LLM agents and workflows. It provides cost caps, approval gates, automatic model switching, retries, and rollbacks to ensure safe and reliable agent operation.

The University of Chicago Law School announced a new policy banning phones and laptops for first-year students, sparking debate about AI in education. This article recalls the history of banning portable computers at Harvard Law School 45 years ago, highlighting the cycle of technological fear. The author shares personal experiences, emphasizing how tools change work processes, and questions the rationality of current policies.

An open, rigorous, living measurement of how much AI engines disagree on which B2B tools to trust per category. In the recorded sample, across 16 categories, all eight models named the same single best tool zero times, with a mean pairwise agreement of 44% and Fleiss' kappa of 0.41. The index is updated monthly and provides raw data for reproducibility.

Anthony Albanese will deliver a landmark speech on AI this week as MPs are torn between attracting datacentre investment and protecting the rights of creatives. Anna Funder described herself as a 'victim of crime' due to tech companies using her works for profit.

Sovereign AgentOps Community Edition is an open-source, self-hosted MCP governance server for AI agents, offering Ed25519-signed audit trails, policy enforcement, and offline deployment. It provides 7 demo tools and aligns with EU AI Act requirements, with a commercial Enterprise edition featuring 91 tools and advanced compliance.

A security scoring system for MCP servers that continuously scans for tool poisoning, prompt injection, supply-chain, and credential risks. Each version gets a single score before agents connect. Out of 12,629 scored servers, 45% received an A grade, while 10% are high-risk (D/F).

A study from University of Maryland and Google DeepMind found that AI-generated fiction is easily detectable due to narrative flaws like over-explaining themes, lack of subplots, and clunky moralizing. The researchers developed StoryScope, a detector that analyzes narrative features, and tested it on over 50,000 AI-generated stories. The study highlighted that different AI models have distinct quirks (e.g., GPT overuses dream sequences, Gemini uses character descriptions). The dataset used includes Books3, which is controversial due to copyright issues. The researchers used AI to assist in writing the paper itself.

Telescope Innovations uses self-driving labs (SDL) to automate chemistry, addressing the physical bottleneck in drug discovery. With deployments at Pfizer, KPBMA, and a European pharma company, plus battery materials breakthroughs, the company is positioned as a key Physical AI player.

The article argues for AI that extends human will and judgment, emphasizing distributed knowledge, customization, and decentralized alignment to ensure AI serves diverse human needs.

SearXNG AI Kit is an AI-enhanced command-line interface, Python library, and MCP server for the SearXNG privacy-respecting metasearch engine, supporting over 180 search engines with standalone binaries available for Linux and macOS.

A personal, non-benchmark tier list of AI models for coding and auditing as of mid-2026, covering Anthropic Fable, OpenAI Sol, Mistral, Gemini, and DeepSeek, with commentary on US export controls and European perspectives.

This video delves into the fundamentals of artificial intelligence, including machine learning, neural networks, and deep learning, providing a clear introduction for beginners.

openpilot 0.11.1 improves driver monitoring with a VLM-based phone detection model, raises thermal thresholds to reduce blocks, adds lateral maneuver reports, and expands car support. The new DM model reduces false positives and better detects active phone use. Thermal changes cut blocked devices by ~90%. New lateral reports aid steering tuning. Bug fixes and new car ports for Acura MDX and Rivian are included.

AnyFile Translator is an AI-powered assistant for Google Chat that translates documents, web links, and messages while preserving original formatting. It supports over 100 languages, offers AI content writing, and ensures data privacy with encryption and deletion.

AI surveillance systems will soon track all public and private actions, enforce rules instantly, and create chilling effects that suppress dissent, creativity, and social progress. The article discusses examples from China and the US, the mechanisms of chilling effects, and calls for policy interventions.

Experts suggest that careers in teaching, hotels, and law may be less affected by AI. While the technology is still evolving, insights from industry professionals offer guidance on future-proofing your career.

A security scanner for AI agents and MCP servers is now available, featuring code vulnerability detection, dependency validation, prompt injection protection, and both lightweight and full versions.

Code Airlock is a lightweight wrapper around Docker Sandboxes that lets coding agents like Claude Code, Codex, and OpenCode run in disposable microVMs with a read-only host repo, enabling safe unattended operation and easy review of changes as git commits.

Omar Yaghi, 2025 Nobel Prize in Chemistry winner, has left the US to lead an AI-driven research center at Tsinghua University in China, aiming to accelerate materials design and synthesis to address environmental challenges like water scarcity and carbon neutrality.

AgentKindergarten is an open-source tool that lets you remotely monitor and interact with AI coding agents from your phone or browser, with real-time terminal streaming, dev server previews, and alert handling. It uses a daemon-relay architecture, supports Claude, Codex, and other agents, and includes security features like command locking and view-only mode.

Pixailer is an AI tool that lets you upload individual photos and describe a scene to generate realistic group photos of up to 8 people in seconds. It offers multiple AI engines (Google Gemini and OpenAI GPT-Image), supports prompts in several languages, and uses a credit-based payment system with no subscription. Privacy is prioritized: images are not used for training and are deleted after generation.

Ant Group's Robbyant has released LingBot-VA 2.0, a causal video-action foundation model designed natively for physical AI. Unlike previous models that fine-tune video generators, this model is pretrained from scratch with a causal DiT backbone, semantic tokenizer, and sparse MoE architecture. Key innovations include Foresight Reasoning for asynchronous control achieving up to 225 Hz, multi-chunk prediction for faster training, and co-training of multiple objectives. On RoboTwin 2.0, it achieves 93.6% average success across 50 tasks.

Krbn is a web engine for non-photorealistic, pencil-style rendering of 3D scenes to SVG. It derives strokes from geometry rather than rasterizing pixels, supporting exact silhouettes, hidden lines, hatching, and more. Written in TypeScript and MIT-licensed, it was developed with AI assistance.