AI Products news and analysis on Web Pulse: 18037 curated articles tracking the latest AI Products developments, tools, and research, updated continuously from vetted sources.
FetchSandbox open-sourced a repository containing intentionally planted security bugs, including a privilege-escalation flaw in a Descope-based agent authentication gateway. The bug allows a read-only agent key to reques…
Dubai-based physical AI startup Ahoy is relocating its R&D hub to Montréal, planning to hire 200 people for research, engineering, and go-to-market roles. The company, which develops AI infrastructure for physical-world …
Cursor, the AI-powered code editor company, is developing a general-purpose AI agent called Sand to compete with Anthropic's Claude and OpenAI's ChatGPT in workplace productivity. The tool, currently in internal testing …
A developer benchmarked the real cost of processing 1,000 invoices per month across five document-AI APIs, revealing hidden minimums and per-attempt billing that inflate effective rates. AWS Textract, Google Document AI,…
Soulless Music has launched an AI-music detector that identifies AI-generated tracks on Spotify and Deezer, exposing artists using tools like Suno and Udio. The project aims to help listeners distinguish human-made music…
Aweb.ai launched Library, an open-source tool that lets teams create, manage, and evolve AI agent profiles as versioned files. The platform enables teams to adopt blueprints, run agents in tmux, and approve profile chang…
Enterprise knowledge is becoming more valuable than AI models as organizations realize that model choice matters less than access to trusted, organized internal data. Retrieval-Augmented Generation (RAG) systems that gro…
Anthropic added a built-in web browser to its Claude desktop app, enabling developers to preview, test, and debug web applications without leaving the AI interface. The feature, part of Claude Code, aims to transform the…
Meta launched Muse Image, an AI image generator integrated into Instagram and WhatsApp that automatically opts in all public Instagram accounts, using their photos as training data without notification or consent. The to…
Voicebox, a free and open-source AI voice studio, launches as a local-first alternative to ElevenLabs and WisprFlow, enabling voice cloning, speech generation in 23 languages, and dictation with full privacy. The app run…
Onnes, a digital twin for dilution refrigerators, uses AI and LLM agents to improve fault diagnosis in quantum computing hardware, achieving 100% recall and near-perfect classification accuracy with minimal training data…
Meta shares surged nearly 15% last week, their best performance since early 2024, driven by new AI product launches and a strategy to fully automate ad creation by 2026. The company's massive AI infrastructure spending, …
Citi reiterated a Buy rating on Nvidia with a $300 price target after speaking with the company's investor relations team, confirming its AI product roadmap remains on track amid rumors of delays. Nvidia also plans a $20…
SpaceX agreed to acquire Anysphere, the company behind the Cursor IDE, for $60 billion in an all-stock deal on June 16, 2026. The acquisition raises concerns about Cursor's model neutrality, as SpaceX's xAI division may …
OpenAI's GPT-5.6 Sol (max) achieved the highest Presentation Elo score in Artificial Analysis's AA-Briefcase benchmark, outperforming competitors like Anthropic's Claude Fable 5 (max) in presentation quality. The model a…
SpaceXAI released Grok 4.5 on July 8, claiming it matches Claude Opus 4.8 in capability while using far fewer tokens. Independent testing by Artificial Analysis confirmed Grok 4.5 uses over 60% fewer output tokens per ta…
Drafted released V2 of its AI home-design tool on Friday, adding room placement controls, partial regeneration, and exterior customization. The San Francisco startup aims to solve the problem of AI-generated home plans t…
Anthropic announced that Claude Code on desktop now includes an in-app browser, allowing the AI to interact with websites similarly to local dev servers. The feature is sandboxed and configurable, with sessions that can …
Enterprise AI agents are runtime products requiring a control plane, API, release path, rollback story, and ownership for regulated workflows. Focused Labs highlights the NemoClaw Deep Agents Blueprint by LangChain and N…
OpenAI employee Vaibhav Srivastav explained the five reasoning levels of GPT-5.6 Sol, recommending users start with lower levels and scale up as needed. The levels, which differ from GPT-5.5's tiers, include "Light" and …