Tags: AI coding assistant

OpenAI unveiled a substantial update to its Codex AI coding platform, adding multi‑app agents, a built‑in browser, image generation, and early memory functions. The enhancements let developers command specific desktop programs, integrate 111 new plugins, and receive proactive suggestions. The rollout begins with macOS users logged into ChatGPT, with EU and UK releases slated for later. While the full super‑app that merges ChatGPT, Codex and a web browser remains in development, the latest release offers a tangible glimpse of OpenAI’s broader vision for a unified desktop AI experience. Weiterlesen

OpenAI announced a sweeping upgrade to its Codex coding assistant, adding background desktop control, memory recall, image generation and a suite of new plug‑ins. The changes let the tool run parallel agents on a user’s Mac, browse the web, and integrate with over a hundred apps. The move, positioned as a direct challenge to Anthropic’s Claude Code, also introduces a pay‑as‑you‑go pricing tier for enterprise customers. Weiterlesen

Anthropic has launched an auto mode for its Claude Code tool, allowing the AI to act on users' behalf while reducing the risk of unwanted actions. The feature flags and blocks potentially risky operations, prompting the model to retry or request user intervention. Currently available as a research preview for Team plan users, Anthropic plans to extend access to Enterprise and API users in the coming days. The company emphasizes that the tool remains experimental and recommends use in isolated environments. Weiterlesen

Anthropic has added a new Code Review capability to its Claude Code AI coding assistant. The feature automatically analyzes pull requests, flags bugs, and supplies actionable feedback through a high‑signal overview comment and inline notes. It scales its multi‑agent review process based on the size and complexity of the change, typically completing a review in about 20 minutes. While the tool costs more than lightweight alternatives, Anthropic offers caps and dashboards to help manage expenses. Early internal testing shows a surge in substantive review comments, and the feature is now rolling out to Claude for Teams and Enterprise subscribers in a research preview. Weiterlesen

Anthropic has begun rolling out Voice Mode for Claude Code, its AI‑powered coding assistant. The feature, announced by engineer Thariq Shihipar on X, is initially available to roughly 5% of users with a wider release planned in the coming weeks. Voice Mode lets developers toggle a /voice command and speak instructions such as “refactor the authentication middleware,” allowing a more hands‑free, conversational coding workflow. While details on usage limits or third‑party voice providers remain unclear, the launch follows Anthropic’s earlier Voice Mode for its standard Claude chatbot and comes amid fierce competition from GitHub Copilot, Cursor, Google and OpenAI. Recent reports show Claude Code’s revenue surpassing $2.5 billion and weekly active users doubling since the start of the year, while the company’s mobile app surged to the top of the U.S. App Store after refusing Department of Defense usage. Weiterlesen

Anthropic has introduced Remote Control for Claude Code, allowing developers to monitor and steer coding tasks from a mobile device. The feature creates a temporary link that mirrors the local session on a phone or web interface, while keeping all files and execution on the original machine. Security relies on one‑time access tokens that expire when the session ends. Remote Control is currently available as a research preview for Claude Max subscribers, with broader rollout planned for other plans. Weiterlesen

Amazon Web Services experienced a 13‑hour outage in December after its AI coding assistant, Kiro, deleted and recreated an environment it was working on. The incident, which affected parts of mainland China, was traced to a human error that gave the bot broader permissions than intended. Amazon says the outage was limited and emphasizes that the root cause was the human‑driven permission issue, not the AI itself, and notes that safeguards and training have been added following the event. Weiterlesen

Amazon Web Services experienced two incidents involving its AI coding assistant Kiro, which were attributed to user error and permission issues rather than flaws in the AI itself. The first incident, described as an "extremely limited event" in mainland China, affected a single service, while the second had no impact on a customer‑facing AWS service. Following the incidents, AWS introduced mandatory peer reviews, staff training, and tighter access controls. The company continues to push for broader AI adoption among developers, targeting 80 percent usage at least once a week, while some employees remain skeptical of the technology’s overall utility. Weiterlesen

OpenAI has introduced the Codex app, a macOS‑only desktop tool that lets software developers orchestrate multiple AI coding agents. The app supports parallel workflows, background tasks, and reusable automations, allowing developers to run code generation, reviews, and scheduled jobs without leaving their local environment. Early users note the ability to manage separate worktrees and threads, reducing the need to switch between terminals, IDEs, and cloud consoles. While the launch is limited to macOS, the feature set signals a shift toward AI agents acting as collaborative teammates in the software development process. Weiterlesen

Cursor, the AI coding assistant, announced it has acquired Graphite, a startup that uses artificial intelligence to review and debug code. While the deal terms were not disclosed, reports indicate Cursor paid well above Graphite's last $290 million valuation. The acquisition adds Graphite's "stacked pull request" technology to Cursor's Bugbot product, aiming to streamline the transition from AI‑generated code to production. The move reflects a broader trend of AI tools targeting the code‑review market, with competitors like CodeRabbit and Greptile also raising sizable funding. Shared investors such as Accel and Andreessen Horowitz link the two companies, and the deal follows Cursor’s recent spree of acquisitions, including Growth by Design and talent from Koala. Weiterlesen

At Fortune’s AI Brainstorm conference, Cursor co‑founder and CEO Michael Truell said the company has no near‑term IPO plans and is instead concentrating on expanding its product suite. He highlighted the rollout of home‑grown large language models, a shift to a usage‑based pricing model that passes API fees directly to customers, and new cost‑management tools for enterprises. Truell also outlined future priorities, including more complex agentic functions such as automated bug fixing and a deeper focus on serving whole development teams. He positioned these moves as a way to stay competitive amid growing AI coding‑assistant rivals. Weiterlesen

Anthropic has launched a new feature that lets Slack users invoke Claude Code directly from their chats. By tagging Claude on a coding‑related message, the system automatically detects the task, pulls context from the thread, and routes the request to Claude Code, which can interact with authenticated code repositories. The functionality is available as a research preview within the existing Claude app for Slack, requiring no additional downloads. The rollout follows Anthropic’s recent release of the Claude Opus 4.5 model, which the company touts as a strong performer in coding tasks compared with competing AI systems. Weiterlesen

Anthropic has introduced a web version of its Claude Code AI coding assistant, featuring a new sandbox that limits file system and network permissions to specific folders and servers. The network isolation works through a Unix domain socket linked to an external proxy that enforces domain restrictions and asks users for confirmation on new requests. Users can customize proxy rules, allowing the agent to fetch npm packages from approved sources without unrestricted internet access. While the changes reduce approval steps and improve security against prompt injection, Anthropic notes that developers must still prioritize code review to catch potential mistakes. Weiterlesen

Anthropic has introduced a web‑based version of its AI coding assistant, Claude Code, extending the tool beyond its terminal interface. The new web app is rolling out to subscribers of Anthropic’s Pro and Max plans, letting users access Claude Code via the Claude website or iOS app. The move aims to let developers spin up AI coding agents wherever they work, positioning Anthropic against rivals such as GitHub Copilot, Cursor, Google and OpenAI. The product, which has seen rapid user growth and now generates significant revenue, continues to rely heavily on Anthropic’s own AI models, with most of its code written by the models themselves. Weiterlesen

Google has broadened the reach of its AI coding assistant Jules by launching a command‑line interface and opening a public API. The new tools let developers invoke Jules directly from terminals, CI/CD pipelines, and chat platforms, reducing the need to switch between web pages and code repositories. Jules will continue to run on Google’s Gemini 2.5 Pro model and now supports more scoped, autonomous tasks while still prompting users when it encounters obstacles. The rollout follows recent feature upgrades such as memory, diff‑viewer enhancements, and broader version‑control support, positioning Jules as a deeper fixture in professional developer workflows. Weiterlesen