Full-text search across 25233 articles. Combine with topic and date filters; results sorted by relevance.
A developer compiled a practical field guide to failure modes in LLM and agent systems, drawing on recent research from Anthropic, Cognition, Meta AI, and others. The guide covers model-level, agent-level, and system-lev…
AI startup hoodies from OpenAI and Anthropic are selling for up to $250 on eBay, reflecting a desire for cultural capital and proximity to Silicon Valley's innovation engine as IPOs remain elusive. The phenomenon highlig…
1Password launched Agentic Mode, an integration with Anthropic's Claude that lets the AI log into user accounts without ever seeing the credentials. 1Password injects passwords and MFA codes directly into the target site…
1Password launched 1Password for Claude, a browser integration that gives Anthropic's Claude AI agent secure access to stored credentials without exposing them to the model or memory, requiring user biometric approval fo…
South Korea targets launching a sovereign AI model for cybersecurity by the end of 2026, Science Minister Bae Kyung-hoon announced at a policy briefing led by President Lee Jae Myung. The plan follows U.S. export restric…
A Meta Oversight Board study of 10 major AI models found that chatbots refused politically sensitive requests 34% of the time for authoritarian jurisdictions like China and Saudi Arabia, versus 14% for democracies, revea…
A friend of the author failed a senior AI engineer interview at an unnamed company after being asked what happens when a retriever returns a document contradicting the user's intent, a question that now represents the ne…
Meta's Oversight Board released a study Thursday finding that leading AI models from labs including Anthropic, OpenAI, Meta, Google, and China's DeepSeek are significantly less likely to criticize governments with repres…
The Oversight Board tested 10 commercial large language models from Anthropic, DeepSeek, Google, Meta, OpenAI and xAI and found they were more than twice as likely to refuse requests to create politically critical materi…
The Oversight Board, the independent content moderation body funded by Meta, published a report finding that leading AI models from OpenAI, Meta, Google, Anthropic, and xAI may restrict free expression by responding diff…
Roughly 73% of global organizations have integrated AI into cybersecurity defenses, and nearly all plan to keep investing through 2025, as the AI industry faces rising physical threats and an 'AI vs. AI arms race' with a…
Anthropic security engineer warns that AI-driven vulnerability discovery has triggered a 'vulnpocalypse' — a surge of software flaws that overwhelms traditional patching processes. The author argues that systemic fixes, …
Anthropic's Claude Opus 4.6 and OpenAI's GPT-5.3 were tested across over 50 structured benchmarks in coding, writing, reasoning, creativity, and vision tasks. Claude Opus 4.6 won decisively in production-quality coding a…
XAI's Grok 4.5, launched on July 8, has climbed to the number two spot on the FrontierSWE leaderboard with a score of 4.09, beating Anthropic's Claude Opus 4.8 (4.82) and OpenAI's GPT-5.5 (5.21). The model, priced at $2 …
Quantum Circuit Vision, a new benchmark of 132 quantum circuits across 13 categories, evaluates multimodal AI agents' ability to understand circuit diagrams and generate executable code for Amazon Braket. Anthropic's Cla…
A developer has released an open-source desktop AI companion built on Tauri v2, React, and Rust that stays always on top for side-by-side coding and querying. The lightweight app supports multiple AI providers including …
South Korea will launch a sovereign AI model specializing in cybersecurity by the end of this year, Science Minister Bae Kyung-hoon announced Thursday, citing growing digital security threats and concerns over US export …
Mira Murati's Thinking Machines Lab released Inkling, an open-weight AI model with 975 billion total parameters under an Apache 2.0 license on July 15. The Mixture-of-Experts model processes text, audio, and video with 4…
1Password launched 1Password for Claude, a zero-exposure architecture that lets Anthropic's Claude AI agent use credentials for browser tasks without the secrets ever entering the model's memory or context. The system re…
New Mexico's proposed AI in Government Act would require transparency for AI used in state services and employment, as colleges face faculty backlash over costly AI data contracts and an AI data center proposal in Raton …