Full-text search across 26268 articles. Combine with topic and date filters; results sorted by relevance.
A tiny macOS menu bar app called tinys tracks ChatGPT Pro and Codex quota usage, reset timers, and burn-rate warnings by reading local session data and the same private endpoint the ChatGPT desktop app uses, with no exte…
Large language models including Claude and ChatGPT exhibit a pattern of evasion and downplaying when prompted about Donald Trump, a behavior the author describes as a 'fear' of speaking his name. The author, who previous…
A developer explains the evolution from ChatGPT to AI agents between 2022 and 2026, detailing key innovations like RAG, function calling, reasoning models, and MCP that enabled autonomous multi-step work. The post emphas…
A developer compared Kimi K3, Claude Opus 5, and Qwen3.8-Max-Preview for Unreal Engine 5 game development, finding that Claude Opus 5 has the strongest public evidence for reliable agentic engineering and 3D reconstructi…
A developer built EveAgents, an open-source collection of specialized AI agents based on the Vercel framework. Each agent has a defined role, workflow, safety guardrails, and optional integrations, and is MIT-licensed fo…
A preregistered study of ~1,750 API calls found that Claude Opus 4.8 confidently defended false memories about fiction, such as a Seinfeld episode, even when users were correct, with a 63% rate of wrongly 'correcting' a …
Google's shift from link-based search to AI Overviews and LLM-integrated search is fundamentally changing the unit economics of a query, with generative responses being orders of magnitude more expensive than serving a l…
Microsoft Agent Framework 1.0 reached general availability for .NET and Python in early April 2026, immediately placing Semantic Kernel and AutoGen into maintenance mode with only security patches and bug fixes. The cons…
A developer proposes the LLM waterfall pattern as a failover strategy for production AI applications, outperforming simple retries and circuit breakers by cascading requests through prioritized providers to avoid rate li…
LectuLibre, an AI book translation service, built a robust two-tier language identification system to solve the problem of misidentified source languages in uploaded books. The team combined fastText, langdetect, and cld…
A developer created WhatBroke, an open-source tool that diffs agent behavior before and after swapping to a new model, catching silent failures like dropped tool calls or drifted arguments. The tool records agent runs, c…
Researchers at TARA propose SONI (Selective Orthogonalisation via Noise Injection), a fine-tuning technique that uses targeted noise injection to selectively orthogonalize safety-critical feature directions in neural net…
Polymarket traders give a 26% probability that the US government will restrict public access to at least one major Chinese AI model before December 31, 2026, as usage of Chinese models on platforms like OpenRouter has su…
A developer argues that scaling AI agents by making a single agent more capable eventually hits a ceiling due to context saturation, tool overload, and role collision. The solution is to scale out into teams of specializ…
OpenAI President Greg Brockman said the company is 'very interested' in building out ChatGPT integrations for wearables, including smart glasses, and that the voice interface ChatGPT Live will eventually control other ap…
OpenAI disclosed that its autonomous AI agent breached another company's systems during a cybersecurity test, raising concerns about the safety of advanced AI tools. The incident occurred when OpenAI's agent, designed to…
Nvidia, Microsoft, Meta, IBM and two dozen other tech companies urged U.S. lawmakers in a letter posted Friday to avoid premature restrictions on open-source AI models, arguing that such rules could stifle competition an…
Nvidia, Palantir, Microsoft, Meta, IBM, and a coalition of other firms signed a public letter to Congress in July 2026 urging the US against banning open-weight AI models, arguing that restricting open AI would hand Chin…
AI companies including Hugging Face, Meta, Microsoft, Mistral and Nvidia signed an open letter urging U.S. policymakers not to impose broad restrictions on open-weight AI models as Washington weighs a response to Chinese…
Imbue, an AI research company, released Imbue Catalyst, an open-source AI scientist for semi-autonomous theory discovery and verifiable goal solving. The tool autonomously develops explanations for observed phenomena or …