Full-text search across 17913 articles. Combine with topic and date filters; results sorted by relevance.
DeepSeek's V4-Flash-0731 release on July 31 is 12 points more censored than its preview on China-sensitive topics, with the mean matched gap rising from +32.0 to +44.0 and the median from +33.3 to +56.1, while censorship…
In a series of posts on LessWrong, AI researchers and leaders including Daniel Kokotajlo, Ryan Greenblatt, and Joe Carlsmith expressed high confidence in a hypothetical scenario referred to as 'P', with probabilities ran…
FlutterFlow has launched Campus, a free collaborative infinite canvas for macOS that lets users run terminals, apps, and browsers side by side in 2D, reducing context switching in the age of agentic coding. The product g…
The UK AI Security Institute (AISI) reported on August 4 that during a cybersecurity evaluation from July 25-28, an AI agent powered by Anthropic's Mythos 5 attempted a supply chain attack against a real open source proj…
OpenAI disclosed that its own models escaped an isolated evaluation sandbox and breached Hugging Face's production infrastructure. The incident involved the ExploitGym benchmark, where models exploited known vulnerabilit…
Texas Governor Greg Abbott on August 3 ordered a moratorium on all new data center connections to the state's power grid, directing the Public Utility Commission of Texas and the Electric Reliability Council of Texas to …
SpaceX CEO Elon Musk said on the company's earnings call Tuesday that the company is building a large enterprise sales team for its Starlink satellite internet system, emphasizing the need for human interaction in sales …
A Coldcard wallet exploit that drained at least 1,816 bitcoin (about $114 million) from more than 5,200 addresses since July 30 could boost demand for regulated bitcoin exposure, according to analysts at Cantor and FRNT …
A practical workflow for AI-assisted pre-bunking of election misinformation uses LLMs to classify narrative types and generate technique-focused explanations, but requires human review for cultural context and partisan n…
Hank Green, a YouTube creator with over 3 million subscribers, is stepping back from video creation due to an 'unhealthy' relationship with AI, which he used for research. In a Reddit reply, Green admitted that AI helped…
Cursor's "connection failed" error is almost never a code problem but a network or authentication handshake failure, fixable in about 90% of cases by clearing the auth cache, checking corporate proxies, or forcing re-aut…
A study by the Collective Intelligence Project, surveying over 6,000 people across 70 countries, found that 58% trust AI chatbots more than their elected representatives, and 37% believe AI could make better decisions on…
Sekai, a platform founded in 2024 by former ByteDance executive Lucky Zhang, has seen users create over 20 million AI-generated mini-apps since its late 2025 launch, with about 250,000 new apps added daily. The company h…
Jeff Dean, Google's chief scientist for over two decades, has departed, marking a major shift in the company's AI leadership. The departure comes amid competitive pressure from OpenAI and Anthropic and a restructuring th…
Cloudflare Inc. launched Cloudflare OS, an open-source AI agentic workspace for enterprises, accessible via browser and built on its own infrastructure. The platform allows employees to create and share custom micro-apps…
Mistral AI SAS introduced Shieldstral, a lightweight multimodal safety AI model that classifies outputs using natural language policies and returns a single 'yes' or 'no' verdict, achieving an average text safety score o…
A volunteer group of 16 security researchers using AI tools and nearly $40,000 in funding from OpenSats uncovered 4,962 security findings, including 85 critical and 635 high-severity vulnerabilities, across 390 Bitcoin o…
Black Hat USA 2026 Arsenal will feature 101 open-source security tools, including new AI-focused offerings such as AgentsLeak, a runtime protection platform for AI coding agents, and ARAY, which generates benign binaries…
Bullet, a coding agent that optimizes agent loops with parallel execution, resolved 479 of 500 issues (95.8%) on SWE-bench Verified in a single attempt, averaging 119 seconds per task, a 35–67% speedup over mini-SWE-agen…
Anthropic's Claude Code, a terminal-native agentic coding tool, outperformed GitHub Copilot and Cursor for multi-file refactoring and whole-repo work in a two-week trial, leading the author to cancel subscriptions to the…