Large language model (LLM) news — GPT-4, Claude, Gemini, Llama, Mistral and the latest research on training, fine-tuning, RLHF, and deployment of LLMs.
A Hacker News user asked the community what they have built using Claude Managed Agents, seeking real-world applications and experiences with the AI tool.
A developer has added a local setup harness to the A-Modular-Kingdom (AMK) project that automatically configures local AI agents to connect to the AMK MCP server. The harness includes a bootstrap script and a thermal run…
The FreeBSD Project has launched a six-month, $250,000 AI-Assisted Vulnerability Discovery Project funded by the Linux Foundation's Alpha-Omega initiative, with support from Microsoft, AWS, Google, Anthropic, OpenAI, and…
Anthropic shut down access to its Fable 5 and Mythos 5 AI models after the White House ordered export restrictions following concerns that the models could be used for cyberattacks and may have been accessed by a group l…
A critical vulnerability chain in LiteLLM, a widely deployed open-source AI gateway, allows attackers starting from a default low-privilege account to achieve full server takeover and code execution. Obsidian Security re…
Anthropic's guide on agentic design patterns categorizes all agentic systems into workflows and agents, where workflows use predefined code paths and agents let LLMs dynamically direct their own processes. The guide emph…
A developer returned to a humbling hangman take-home challenge years later and built a Transformer model that runs in the browser after 16 experiments and 6 failed reinforcement learning attempts.
The European Commission ruled in April that Google's integration of Gemini into Android violates the Digital Markets Act, proposing measures to force Google to give rival AI services equal system-level access, including …
A user observes that AI models like ChatGPT, Gemini, GROK, and Claude frequently hallucinate, leading users to switch between models. The author notes that end users often lack understanding of AI limitations, while rese…
AI safety researcher argues that expensive safety measures can be applied selectively to the <1% of tasks carrying catastrophic risk, making the alignment tax economically viable. The blended overhead of a 100× tax on 0.…
A technical report proposes using a consumer RTX 3090's integrated GPU (iGPU) to run a small language model as a 'Sentinel' for monitoring and validating outputs from the primary GPU-bound model. The approach aims to imp…
The U.S. government imposed export controls on Anthropic's Fable and Mythos AI models after Amazon researchers discovered a security vulnerability triggered by the simple prompt 'fix this code,' which could be exploited …
Meta announced new AI features on Facebook, including 'AI Mode' that synthesizes answers from public posts across its platforms, and AI-powered photo presets for editing. The updates aim to boost engagement and diversify…
A Hacker News user asked why large language models frequently use the em dash (—) in their responses, noting that this punctuation was uncommon in emails before AI. The user cited an example of AI-generated text containi…
A senior backend tech lead and founder of HostingGuru warns that many developers have accidentally committed API keys to Git history, often through .env files or hardcoded keys in AI coding tools. He recommends using git…
A developer built a web tool that checks a user's hardware in the browser, pulls model info from Hugging Face, estimates memory requirements, and returns a simple yes/no/maybe answer about whether a given model can run l…
A developer's field guide categorizes AI agent users by their technical proximity and needs, from non-technical users who want reliable task completion to executives focused on strategic outcomes. The piece highlights ho…
Shrijith Venkatramana, developer of git-lrc, explains how kernel fusion reduces memory bandwidth bottlenecks in LLM inference. By combining multiple GPU operations into a single kernel, intermediate data movement is mini…
Anthropic faces mounting challenges as coding agents like Claude Code prove easy to switch from, with users migrating to OpenAI's Codex after rate limit frustrations. The lack of a moat in coding agents is highlighted by…
The Trump administration forced Anthropic to take its Fable 5 and Mythos 5 AI models offline, blocking access for foreign nationals including employees. The shutdown has intensified global calls for sovereign AI developm…