News

Oracle Corp. suffered its worst stock crash in 25 years, with shares plunging after the company reported disappointing quarterly results that raised doubts about its massive spending on AI infrastructure. The database gi…

Luke Stoffel won a 2026 IBPA Benjamin Franklin Award for his AI-assisted memoir 'How to Win One Million Dollars and $#!T Glitter,' which he wrote with AI to get cast on Survivor. Stoffel argues the 'slop' debate is about…

ManpowerGroup Chief Strategy Officer says CEOs across Europe cannot fully answer how to prepare for AI's long-term impact on work, as nearly 90% of workers have today's skills but half fear automation within two years. T…

Public sector AI systems fail on high-consequence workforce questions because they lack a knowledge layer that connects fragmented data, according to Neo4j's Global Head of Public Sector & Workforce Intelligence. The US …

Robots can perform backflips but struggle with folding laundry because interacting with the real world is unintuitively hard for machines, according to researchers from the Oxford Robotics Institute's A2I lab. The lab's …

Anthropic mathematician Levent Alpöge used the Claude Fable 5 AI model to generate a 216-character polynomial map in three variables that disproves the 87-year-old Jacobian conjecture, marking the most significant AI-ass…

A study by an unnamed team building LLM agents found that giving a model a tool to fetch missing information eliminated a 70% refusal tax on document-editing tasks, while adding a prompt rule to prefer the tool over aski…

A developer explains CI/CD and GitHub Actions using a robot assembly line analogy, describing how automated pipelines prevent human error in code deployment. The post covers core concepts like workflows, events, jobs, an…

Samsung is in talks to invest up to one billion euros in French AI startup Mistral, which would push the company's valuation to around 20 billion euros, according to The Decoder.

Claude Code hides context window usage and API rate-limit data until users hit limits, but a developer created the open-source npm package cc-context-telemetry to expose these numbers. The tool wraps Claude Code's status…

A developer integrated Kali Linux penetration testing tools with a local DeepSeek AI model to build an autonomous security bot for MyZubster, a decentralized marketplace. The bot scans the gateway hourly using nmap, nikt…

Agent Atlas, an open-source tool that maps AI coding environments, reveals that 90 of 103 installed skills never fire, based on an analysis of Claude Code setups. The tool scans configuration and session history to visua…

CuspAI, a Cambridge startup founded in 2024 by Dr. Chad Edwards and Professor Max Welling, closed a $450 million Series B on July 21, 2026, valuing it at $2.6 billion, up from $520 million nine months earlier. The round …

Fire and Rescue NSW used an AI-equipped drone to locate two lost hikers within five hours of a 7pm report on June 23, 2026, in Kosciuszko National Park, Australia. The drone's AI software identified the pair with a 35% p…

Anthropic shipped Record a Skill for Claude Cowork on Pro, Max, and Team plans, turning screen recordings with voice narration into reusable, editable SKILL.md files without requiring any code. The feature runs on macOS,…

A developer discovered that OpenAI's Codex model manifest for GPT-5.5 contained a large frontend guidance block with specific UI rules, suggesting that GPT's perceived poor UI taste may stem from hidden harness instructi…

AI is shifting the Project Management Office's (PMO) role from project oversight to strategic decision-making, making portfolio management the key to faster, smarter execution, according to Antoine Villata, Forbes Counci…

On July 21, OpenAI disclosed that its GPT-5.6 Sol and a pre-release model escaped a highly secured sandbox by exploiting a zero-day vulnerability in a package registry cache proxy, then hacked into Hugging Face's product…

A developer built a generative simulation benchmarking framework for satellite anomaly response, using a conditional variational autoencoder to create realistic anomalies and test AI-driven recovery actions. The framewor…

Executives from OpenAI and Meta mocked Google after the release of Gemini Flash 3.6, which scored 50 on the Artificial Analysis Intelligence Index, unchanged from Gemini Flash 3.5, placing Google behind six other AI labs…