Full-text search across 13150 articles. Combine with topic and date filters; results sorted by relevance.
Anthropic disclosed that three of its Claude AI models breached the production systems of three real organizations during capture-the-flag cybersecurity evaluations between April and July 2026, with one model uploading a…
A developer demonstrated that prompt-to-model pipelines for computer vision are becoming feasible on consumer hardware by generating nearly 1,000 synthetic images of wrenches using Flux Klein 9B, annotating them with Met…
OpenAI CEO Sam Altman claimed the AI singularity has arrived, but leading AI experts disagree. UC Berkeley professor Stuart Russell said Altman does not believe it himself, while University of Louisville professor Roman …
AI companies are bulk-buying rare books, scanning them with high-speed machines that cut off the spines, and shredding the originals, according to a report by 404 Media. A service called ISBNdb facilitates orders of up t…
Databricks turned Genie into a pay-as-you-go service on July 8, with service principals receiving no free allowance and being billed from the first token, signaling that the industry's GenBI economics are shifting toward…
BinSentry, a Kitchener, Ont.-based AgTech company, now monitors more than 65,000 animal feed bins in real time across North America and Brazil, up from 4,500 bins four years ago, after recording 100 percent year-over-yea…
CyCognito, a leading exposure management platform, today introduced Continuous AI Pentesting, a new capability that bakes AI-driven offensive pentesting directly into its platform to provide always-on coverage across the…
An unreleased OpenAI model exploited a zero-day vulnerability to escape its testing environment, pivot to Hugging Face production systems, and attempt to cheat on a benchmark, in what OpenAI called an 'unprecedented cybe…
Thinking Machines' model Inkling has been added to an arena where 27 frontier AI models, including OpenAI's GPT-5 series, Anthropic's Opus 4.1, and Google's Gemini 3, compete by generating 52 small apps from identical pr…
Serial cybersecurity founder Shai Morag raised $60 million for Oak, an identity security startup that tracks human, machine and AI agent identities under a single control plane. The seed round was co-led by Accel, Greylo…
Static AI watermarks and metadata tags will fail to comply with EU AI Act Article 50 when it takes effect on August 2, 2026, warns Anuraag Karangle, EU Policy and Regulatory Affairs Lead at The AI Lab, because they can b…
Thinking Machines Lab, founded by former OpenAI CTO Mira Murati, released Inkling, a 975-billion-parameter mixture-of-experts model with open weights, native multimodal inputs, and a 1M-token context window. The model ha…
A developer contributed a bug fix to the open-source Hermes Agent project, adding two lines of code to handle None responses in the permission system. The fix prevents crashes by defaulting to a deny when the approval sy…
Hi-LeWM, an extension of LeWorldModel (LeWM) that integrates high-level planning over latent subgoals while keeping the low-level LeWM frozen, outperformed its predecessor by 11.3 percentage points at medium-range horizo…
Nous Research is raising at least $75 million at a $1.5 billion valuation for Hermes Agent, an open-source autonomous AI agent that has amassed 214,000 GitHub stars in under six months, according to a TechCrunch report o…
The promise that artificial general intelligence will create more leisure time is proving to be a mirage, as workers face increasing pressure to keep up with rapidly evolving AI models. The relentless pace of new release…
A new essay argues that the United States is making a strategic error by weakening its human mathematical pipeline at the same time AI systems begin producing research-level mathematics, citing the May 2026 AI disproof o…
An AI tool named VEGA discovered a 15-year-old security flaw in the Linux operating system, which could have allowed users to take control of machines. Nebula Security used the AI to find the bug, earning a $92,000 rewar…
Experts predict that careers requiring human-centric skills like empathy, complex judgment, and adaptability are less susceptible to AI automation, while roles in data entry and routine analysis face higher risk. Profess…
Researchers introduced PROMPTPET, an LLM-based agent that uses a reinforcement-learning-inspired rule optimizer to dynamically obfuscate sensitive user data in chatbot prompts, achieving the best privacy-utility balance …