InfoQ Homepage Artificial Intelligence Content on InfoQ
-
Anthropic Introduces Agent-Based Code Review for Claude Code
Anthropic has introduced a new Code Review feature for Claude Code, adding an agent-based pull request review system that analyzes code changes using multiple AI reviewers.
-
Cursor 3 Introduces Agent-First Interface, Moving beyond the IDE Model
Anysphere released Cursor 3, a redesigned interface built from scratch that shifts the primary model from file editing to managing parallel coding agents. The new workspace supports local-to-cloud agent handoff, multi-repo parallel execution, and a plugin marketplace. Community reaction has been divided, with developers questioning cost overhead and the move away from Cursor's IDE-first identity.
-
Claude Code Used to Find Remotely Exploitable Linux Kernel Vulnerability Hidden for 23 Years
Anthropic researcher Nicholas Carlini used Claude Code to find a remotely exploitable heap buffer overflow in the Linux kernel's NFS driver, undiscovered for 23 years. Five kernel vulnerabilities have been confirmed so far. Linux kernel maintainers report that AI bug reports have recently shifted from slop to legitimate findings, with security lists now receiving 5-10 valid reports daily.
-
Anthropic Releases Claude Mythos Preview with Cybersecurity Capabilities but Withholds Public Access
Anthropic has introduced Claude Mythos Preview, its most advanced AI model, improving significantly in reasoning, coding, and cybersecurity. Unlike previous releases, it will not be publicly available. Access is limited to a consortium of tech companies through Project Glasswing. Internal tests revealed the model's ability to discover critical security flaws effectively.
-
AAIF's MCP Dev Summit: Gateways, gRPC, and Observability Signal Protocol Hardening
The MCP Dev Summit North America 2026, held on April 2-3 at the New York Marriott Marquis, gathered about 1,200 attendees. Hosted by the Linux Foundation's Agentic AI Foundation, discussions focused on the Model Context Protocol's evolution and enterprise adoption, particularly by Amazon and Uber, emphasizing security, interoperability, and scaling for production.
-
Google Brings MCP Support to Colab, Enabling Cloud Execution for AI Agents
Google has released the open-source Colab MCP Server, enabling AI agents to directly interact with Google Colab through the Model Context Protocol (MCP). The project is designed to bridge local agent workflows with cloud-based execution, allowing developers to offload compute-intensive or potentially unsafe tasks from their own machines.
-
Cloudflare and ETH Zurich Outline Approaches for AI-Driven Cache Optimization
Cloudflare and ETH Zurich highlight how AI-driven crawler traffic challenges traditional caching in CDNs and databases. They propose AI-aware strategies including separate cache tiers, adaptive algorithms, and pay-per-crawl models to balance performance for human users and AI services while maintaining cache efficiency and system stability.
-
Istio Evolves for the AI Era with Multicluster, Ambient Mode, and Inference Capabilities
The Cloud Native Computing Foundation (CNCF) has announced a major evolution of Istio, introducing new capabilities aimed at making service meshes “future-ready” for AI-driven workloads.
-
Dynamic Languages Faster and Cheaper in 13-Language Claude Code Benchmark
A 600-run benchmark by Ruby committer Yusuke Endoh tested Claude Code across 13 languages, implementing a simplified Git. Ruby, Python, and JavaScript were the fastest and cheapest, at $0.36- $0.39 per run. Statistically typed languages cost 1.4-2.6x more. Adding type checkers to dynamic languages imposed 1.6-3.2x slowdowns. Full dataset available on GitHub.
-
Helidon 4.4.0 Introduces Alignment with OpenJDK Cadence and Support via Java Verified Portfolio
Oracle has released version 4.4.0 of Helidon, their microservices framework, featuring alignment with the OpenJDK release cadence, support via the new Java Verified Portfolio, new core capabilities, and agentic AI support for LangChain4j.
-
GitHub Will Use Copilot Interaction Data from Free, Pro, and Pro+ Users to Train AI Models
GitHub will use Copilot interaction data from Free, Pro, and Pro+ users to train AI models starting April 24, opting in by default. Collected data includes code snippets, inputs, outputs, and navigation patterns from active sessions, including private repos. Business and Enterprise tiers are excluded. Community concerns include dark patterns, IP exposure, and GDPR compliance.
-
Pinterest Deploys Production-Scale Model Context Protocol Ecosystem for AI Agent Workflows
Pinterest engineering teams have deployed a production-ready Model Context Protocol (MCP) ecosystem that allows AI agents to automate complex engineering tasks and integrate diverse internal tools. Domain-specific MCP servers, a central registry, and human-in-the-loop approval improve security, governance, and developer productivity while saving thousands of hours per month.
-
Cloudflare Launches Dynamic Workers Open Beta: Isolate-Based Sandboxing for AI Agent Code Execution
Cloudflare has released Dynamic Worker Loader into open beta, offering V8 isolate-based sandboxing for AI-generated code execution. The company claims isolates start in milliseconds, using megabytes of memory, making them roughly 100x faster and up to 100x more memory-efficient than containers. The feature builds on Cloudflare's Code Mode approach.
-
QCon London 2026: Team Topologies as the ‘Infrastructure for Agency’ with AI
At QCon London 2026, Matthew Skelton argued that AI success depends on organisational maturity. He highlighted bounded agency, security, and stewardship as key to managing AI agents. By using Innovation and Practices Enabling Teams, companies can drive knowledge diffusion and optimise internal processes to see real-world returns on their AI investments.
-
Google Unveils AppFunctions to Connect AI Agents and Android Apps
In a move to transform Android into an "agent-first" OS, Google has introduced new early beta features to support a task-centric model in which apps provide functional building blocks users leverage through AI agents or assistants to fulfill their goals.