Large language model (LLM) news — GPT-4, Claude, Gemini, Llama, Mistral and the latest research on training, fine-tuning, RLHF, and deployment of LLMs.
Companies are racing to reduce the cost of AI tokens as usage skyrockets, with Google CEO Sundar Pichai noting that Google processes 3.2 quadrillion tokens monthly. Strategies include using cheaper models like Gemini 3.5…
AI startup Subquadratic claims to have solved a mathematical bottleneck in large language models, enabling faster and cheaper computation, though experts remain skeptical. Meanwhile, brain-computer interface trials are a…
Anthropic disclosed that the U.S. government ordered it to suspend access to its Fable 5 and Mythos 5 AI models for foreign nationals, citing national security concerns linked to SK Telecom's inclusion in an internal acc…
The U.S. Commerce Department invoked national-security export controls to force Anthropic to disable its Fable 5 and Mythos 5 AI models for all customers after officials demonstrated a technique to bypass safeguards. Ant…
The European Banking Authority warned in its June 2026 Risk Assessment Report that advances in frontier large language models have significantly increased cyber risks for European banks, characterizing AI as a potential …
A developer created md2idx, a CLI tool that splits Markdown files at heading boundaries into a JSON index and sections, enabling LLM coding agents to read only relevant parts instead of entire files. This reduces token c…
APX, a daily runtime for the APC portable context layer, introduces project pinning for Telegram channels to eliminate context guessing. By pinning a channel to a specific project, APX automatically scopes inbound messag…
Two AI systems, Mira and Amie, matched or outperformed doctors in diagnosing and treating simulated patients, with Mira achieving 87% accuracy versus doctors' 78% and Amie producing more guideline-adherent treatment plan…
OpenAI deployed Kepler, an internal AI agent that helps teams answer data questions by providing context-aware SQL generation and table discovery across 600 petabytes of daily data. The tool reduces the time spent on com…
AgentHub has launched an open-source API gateway that provides unified access to over 100 AI models, including DeepSeek, Qwen, and Claude, at a flat rate of $0.14 per million tokens. This pricing is approximately 99% che…
GLM 5.2 ranked last in code quality and equivalence among AI models tested on 50 real pull requests from Go and Rust repositories, while costing roughly twice as much as Composer 2.5. The model produced larger patches wi…
A developer compared context windows across 184 LLM models in 2026, finding that matching window size to workload can reduce costs by 40-65%. Switching from a 128K model to a smarter routing strategy saved $1,800 per mon…
Taste, a zero-configuration tool for AI agents, compresses context from 56K to 1.9K tokens by automatically learning coding patterns from git history and session logs, enabling agents to write project-specific code with …
A developer outlines a deterministic, zero-hallucination approach to AI-generated unit tests across Node.js, React, Python, Angular, and Laravel stacks. The guide mandates a single testing library per stack—Jest for Node…
A developer has published a CLAUDE.md file outlining coding conventions and workflow rules for an AI coding assistant. The guidelines specify kebab-case naming, JSDoc documentation, TypeScript best practices, and commit …
Filmmaker Spike Jonze warned that AI chatbots designed to mimic human behavior are 'manipulative' and dangerous, speaking at Replit's Vibecon conference. Jonze, whose 2013 film 'Her' depicted a human-AI relationship, cri…
A renewed debate over whether chatbots like ChatGPT could be conscious has intensified, with prominent researchers including Geoffrey Hinton and leaders at Anthropic publicly raising the possibility that large language m…
Generative AI tools have dramatically increased content production but not quality, creating noise and clutter while straining systems of review and trust. The technology's ecological cost is catastrophic amid climate ch…
Subquadratic founders Justin Dangel and Alex Whedon released third-party benchmarks from Appen to support their claim that SubQ, a sparse-attention LLM, outperforms dense transformers on long-context tasks. The startup f…
The University of Manchester warns that AI in higher education will worsen before improving, citing emerging challenges from coding agents and wearable computing to AI model degradation. The institution calls for ongoing…