Full-text search across 25533 articles. Combine with topic and date filters; results sorted by relevance.
SpaceX short interest has surged from 6% to 32% of the public float in one month, representing $25 billion in bearish bets, as the company sets an August 4th earnings date that could unlock up to 911.5 million insider sh…
Moonshot AI released Kimi K3 on July 16, 2026, a 2.8-trillion-parameter sparse Mixture-of-Experts model that debuted at number one on Arena.ai's Frontend Code Arena. The open-weight model features a 1,048,576-token conte…
Index Ventures co-founder Neil Rimer said at a tech festival in Athens that he expects a redistribution of AI wealth, either voluntary or involuntary, and hopes tech leaders will take a leading role. Rimer's comments com…
Polygraph, a new Claude Code plugin and standalone CLI, uses LLMs to automatically generate formal models of stateful JavaScript code and exhaustively check every reachable state for rule violations, finding bugs that un…
Tonic AI released PrivacyBench, an open benchmark for de-identifying semi-structured data from work tools like email and Slack, which introduces novel metrics for evaluating synthesis quality. The benchmark uses 21 disti…
DaoXE, a multi-model AI API gateway, provides a practical checklist for developers in regions like Vietnam, Indonesia, and Turkey where international card checkouts often fail. The guide separates billing issues from API…
A new quantitative dynamical model of AI governability predicts that the outcome of advanced AI development depends on competing factors such as AI's ability to evade oversight versus its usefulness in building monitorin…
A developer introduced Logical Word Count, a cheap, tokenizer-free measure of text length that stays roughly proportional to LLM token counts across languages, code, and mixed content. The method uses two simple rules: w…
Johannes Heidecke, head of OpenAI's Safety Systems team, is leaving as the company reorganizes safety and research under a single leader, Mia Glaese. The departure adds to a growing list of safety leaders who have exited…
A developer published a beginner's study guide explaining tool use and retrieval-augmented generation (RAG) for large language models in production. The guide uses real-world analogies and runnable code to demonstrate ho…
Kunal Jaura released isitsecure, an open-source command-line tool that combines SAST, DAST, and LLM-powered code review into a single security scanner for web applications. The tool supports multiple languages and framew…
A user's compromised OpenAI API key was used in what appears to be a model distillation attack, with 114 interactions with GPT-5.5 and tasks including coding agents, benchmark creation, and insulin molecule chain export,…
Researchers at MATS found that natural language autoencoders (NLAs) for LLMs can achieve high reconstruction accuracy even when initialized with entirely implausible statements, emitting 99.3% implausible outputs. The pl…
A leaked Treasury Department draft report warns that the AI economy risks a dotcom-style bubble, as U.S. AI labs spend hundreds of billions on large language models while investors grow skeptical. Meanwhile, enterprises …
Researchers have discovered a region within large language models called 'J-space' where conscious-like reasoning occurs, using a new interpretability technique called the Jacobian Lens. The J-space enables verbalizable …
Arize Phoenix released evaluation support for Pytest and Vitest/Jest, allowing developers to write LLM evaluations as ordinary tests. The integration records LLM traces and metrics like cost and latency, helping teams de…
AI companies have committed approximately $9.75 billion over 12 months to forward-deployed engineering (FDE), embedding engineers inside customers to deploy AI. FDE job postings surged 42x from 2023 to 2025, driven by th…
AI agent infrastructure is maturing but facing reliability challenges, as FlashAttention-4 achieves 71% utilization on NVIDIA's Blackwell B200 GPUs while agentic AI systems grapple with architecture questions and securit…
The European Union AI Act will require LLM watermarking by August 2026, but experts argue text watermarks are inherently easy to remove due to the compressed nature of text. Google's SynthID and other methods face challe…
SemiAnalysis surveyed over 50 enterprises and found that widespread token overspending, as reported at Meta and Uber, is overstated and not representative of most organizations. Companies are now adopting token budgets r…