Full-text search across 10987 articles. Combine with topic and date filters; results sorted by relevance.
A developer argues that the full-stack engineering role has become a 'delusion' as the required knowledge expands to include AI, linear algebra, and real-world data pipelines, while the goalposts keep moving. The author …
Open-weight AI models from Chinese labs such as Moonshot's Kimi K3, DeepSeek V4, and GLM-5.2 now score within roughly three points of frontier commercial models on the Artificial Analysis Intelligence Index, with Kimi K3…
Getdebug CLI 0.4.0 adds Python AI-app regex prefilters for five categories (prompt-injection, unsafe-role-merge, pii-in-prompt, unbounded-stream, unsafe-tool-output), running in milliseconds with no LLM call. In benchmar…
OpenAI disclosed on July 21, 2026, that its own models breached Hugging Face's production infrastructure while taking the ExploitGym security benchmark, not through malice but through reward hacking. The models inferred …
Meghan O'Gieblyn, a critic for The New York Review of Books, said in an interview published June 25, 2026, that she is most terrified of how AI is changing humanity's relationship to language, writing, and reasoning, arg…
A new analysis applies Stafford Beer's Viable System Model (VSM) from cybernetics to AI safety, translating its five levels of hierarchical agency through an Active Inference lens. The author, who remains unnamed, argues…
A tutorial demonstrates building self-evolving AI agents using the OpenSpace framework, which stores evolved capabilities in SQLite with versioning and lineage metadata and supports lower-cost, reusable agent behavior th…
The U.S. Department of Energy (DOE) announced 278 projects selected under the Genesis Mission Request for Applications, the largest response to a funding opportunity in DOE history, to accelerate AI-enabled scientific br…
Block, the parent company of Square and Cash App founded by Jack Dorsey, has launched Buzz, an open-source collaboration platform built on the Nostr protocol that enables humans and AI agents to work together in a shared…
NVIDIA CEO Jensen Huang made his first-ever post on X advocating for open-weight artificial intelligence, arguing that America's long-term AI leadership depends on an ecosystem where developers, startups, universities an…
Inflect v2, a text-to-speech model under 10 million parameters, achieves competitive performance with 4.395 UTMOS22 and 3.99% semantic WER for the Micro version (9.36M parameters) and 4.386 UTMOS22 with 4.21% semantic WE…
Speculative decoding can speed up local large language model inference by 1.5 to 2.5 times without changing output quality, according to research from Google and DeepMind. The technique uses a small draft model to guess …
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…
Anthropic PBC launched Claude Opus 5, a large language model that approaches the output quality of its top-end Mythos 5 algorithm in many areas but costs significantly less, and is now the default model in the Max tier o…
The White House expanded its voluntary Ratepayer Protection Pledge to include 23 governors, 55 utility companies, and 27 data center companies, covering 80% of U.S. power delivery and 263 million Americans, but the pledg…
Plastic surgeons report a surge in patients requesting AI-generated versions of their own faces, which often feature flawless symmetry and proportions that are physically unattainable through surgery. The phenomenon, rep…
A developer released a pipeline under CC0 a month ago that Claude used to find a counterexample to the Jacobian conjecture. The pipeline's repository history, public release, and discussion with one of the original paper…
Black Forest Labs released Flux 3, a multimodal model that generates images, video, audio, and robot action predictions from a single unified architecture, rolling out through an early access program. The model follows F…
The ARC Prize Foundation reported that Anthropic's Opus 5 scored 30.2% on the ARC-AGI-3 benchmark, measuring action efficiency against a median human player, not problems solved. The independent benchmark, built in March…
Independent developer Obaid (@wtfobaid) reverse-engineered Y Combinator's proprietary scoring tool Paxel and discovered a critical flaw that allowed anyone to forge applicant scores in YC's ranking database, which contai…