Agents drive product launches; Regulation targets safety

Anthropic / Claude ecosystem

Agents for financial services and insurance

Anthropic has released ten pre-built agent templates designed for financial services tasks, including pitchbooks, KYC screening, and month-end close. These templates are integrated with Claude Cowork, Claude Code, and Microsoft 365 applications.

Inside Claude Code Auto Mode: Anthropic’s Autonomous Coding System with Human Approval Gates - InfoQ

Anthropic has launched Claude Code Auto Mode, an autonomous multi-step coding system that includes human approval gates. This new feature aims to reduce manual intervention in coding tasks while maintaining safety checkpoints through two-stage classification and subagent workflow validation.

Dun & Bradstreet Brings Risk & Compliance Workflows to Anthropic's Claude

Dun & Bradstreet (D&B) has integrated its commercial risk data directly into Anthropic's Claude via an MCP server. This integration enables automated KYC/KYB onboarding workflows, eliminating manual steps and siloed tools.

Frontier model providers

OpenAI turns its sold-out GPT-5.5 party into a monthlong Codex giveaway for 8,000 developers | VentureBeat

OpenAI has expanded Codex rate limits tenfold for over 8,000 developers who applied for its sold-out GPT-5.5 party. This move aims to drive sustained trial usage and future paid subscription upgrades after the June 5 expiration of the boosted access.

Google Is Building an AI Agent That Could Be Its Answer to OpenClaw - Business Insider

Google is reportedly developing an internal AI agent named 'Remy,' which aims to directly compete with OpenAI's OpenClaw. Remy is designed to offer autonomous task execution seamlessly integrated across Google's suite of services.

AI developer tooling & infrastructure

Opsera and Cursor Partner to Embed Autonomous AI Agents Directly into AI-SDLC Workflows for Next-gen AI-Driven Development

Opsera and Cursor have partnered to embed autonomous AI agents directly into the Cursor IDE. This integration aims to align high-speed code generation with enterprise security, compliance, and architectural standards.

Cloud & platform providers

Copilot Cowork: From conversation to action across skills, integrations, and devices

Microsoft has expanded Copilot Cowork from desktop to iOS/Android mobile platforms and introduced reusable skills and cross-system integrations. This enables AI to execute tasks autonomously across Microsoft 365 and various third-party tools.

Introducing the agent quality loop: AgentCore Optimization now in preview | Artificial Intelligence

AWS has announced the preview of AgentCore Optimization, a new service that automates AI agent quality improvement. This is achieved through production trace analysis, batch evaluation, and A/B testing, designed to replace manual prompt tuning cycles.

Google’s Agentic Data Cloud to power ‘systems of action’ | Computer Weekly

Google is repositioning its data platform as an 'Agentic Data Cloud,' serving as a semantic orchestration layer for autonomous AI agents. This includes introducing cross-cloud and on-premises capabilities with enhanced metadata governance.

AWS Introduces Generative AI Model Agility Solution

AWS has introduced a Generative AI Model Agility Solution, a framework designed to enable enterprises to seamlessly swap Large Language Models (LLMs) without requiring code changes. This shifts the competitive advantage from frontier models to infrastructure and application layers.

Google Cloud Introduces Event-Driven Webhooks for AI Agents

Google Cloud has introduced Event-Driven Webhooks for its AI Agents, allowing them to shift from pull-based request-response models to push-based reactivity. This enables AI agents to trigger autonomous actions in real-time across more than 150 event sources.

Cloudflare Introduces Flagship: an Edge-Native Feature Flag Service Built on OpenFeature - InfoQ

Cloudflare has launched Flagship, an edge-native feature flag service built on OpenFeature. It evaluates flags sub-millisecond at the edge without external round trips, positioning itself as a native alternative to third-party platforms like LaunchDarkly.

Cloudflare Shares Surge as Platform Expands Beyond Cybersecurity Into AI Agent Infrastructure | NewsCase

Cloudflare is expanding its platform beyond cybersecurity into AI agent infrastructure, enabling autonomous agents to operate on its platform with automated domain registration, subscription management, and payments integration.

Cloudflare warns of AI code review prompt injection

Cloudflare's research demonstrates that AI code reviewers can be bypassed via indirect prompt injection embedded in comments. Detection rates fell to 53% when deceptive comments comprised less than 1% of the file size, showing uneven resilience across various AI models.

AI policy, regulation & governance

AI oversight: Trump admin. will test Google, Microsoft and xAI models

The U.S. government has secured pre-deployment access to evaluate frontier AI models from major vendors including Google DeepMind, Microsoft, and xAI before their public release, under new agreements with the Center for AI Standards and Innovation (CAISI).

The White House reportedly is working on an executive order about AI oversight and access. | The Verge

The Trump Administration is reportedly considering an executive order on AI model vetting following concerns raised by the 'Mythos' incident. However, a working group for this initiative has not yet been formally established.

Colorado lawmakers advance rewrite of 2024 law to regulate artificial intelligence - Colorado Politics

Colorado lawmakers are advancing a rewritten AI regulation bill, Senate Bill 189 (SB-189), to replace the criticized 2024 law. The revised version aims to balance innovation with consumer protection.

ADIA Launches Independent Review Of Privacy Code As AI Advancements Pose 'New Risks' - B&T

The Australian Data and Insights Association (ADIA) has launched an independent review of its Privacy (Market and Social Research) Code. The review aims to address new risks arising from the rapid adoption of AI within Australia's data and insights industry.

Australia, Japan commit to partnership to meet cyber security challenges and strengthen cyber defences - Cyber Daily

Australia and Japan have established a formal strategic cyber partnership. This collaboration aims to strengthen joint defenses, foster cooperation on critical technologies, including AI, and enhance regional resilience in the Indo-Pacific.

Industry & market moves

PYMNTS | FIS and Anthropic Collaborate to Enable Agent-First Banks

FIS and Anthropic have collaborated to launch a Financial Crimes AI Agent, significantly accelerating anti-money laundering investigations from hours to minutes. This partnership plans to expand to additional banking agents deployed via FIS's governed platform.

After Pushback, Amazon Rolls Out Claude Code, Codex to All Employees - Business Insider

Amazon has formally expanded access to Anthropic's Claude Code and OpenAI's Codex across its corporate workforce, following internal complaints. This move signals that AI coding assistants are becoming essential infrastructure for the company.

OpenAI and PwC collaborate to reimagine the office of the CFO

OpenAI and PwC have launched a production collaboration aimed at deploying AI agents across enterprise finance operations. This partnership will use OpenAI's internal finance organization as a validation model for these AI solutions.

Google DeepMind workers in UK vote to unionize amid deal with US military

Google DeepMind's UK workers have initiated a vote to unionize, becoming the first at a frontier AI lab to seek formal recognition. This action is driven by concerns regarding the militarization of AI and the company's existing defense contracts.

NVIDIA and ServiceNow Partner on New Autonomous AI Agents for Enterprises | NVIDIA Blog

NVIDIA and ServiceNow are collaborating to bring governed autonomous AI agents to enterprise desktops with 'Project Arc.' This initiative combines open models, a secure runtime (OpenShell), and ServiceNow's governance framework for production-scale deployment.

ZyG announces $60M Series A Round Led by Accel

ZyG, an AI e-commerce platform founded by ironSource alumni and Israeli cyber experts, has raised a $60M Series A funding round led by Accel. This funding will be used to build an agentic operating system for automating eCommerce scaling for DTC brands.

CopilotKit raises $27 million to build infrastructure for AI agents | Ctech

CopilotKit has raised $27 million in Series A funding, with its AG-UI open standard for AI agent interactions being adopted by Google, Microsoft, Amazon, and Oracle. This positions the startup as a key infrastructure backbone for enterprise AI deployment.

AI product & feature launches

Meta Is Using AI Bone Structure Analysis To Detect And Remove Underaged Users

Meta is deploying AI-powered bone structure and visual analysis to identify and remove underage users from its Facebook and Instagram platforms. This expands their automated age verification systems.

Meta expands AI age technology to EU - RTE

Meta is expanding its AI-powered age-detection system, which identifies underage users through profile analysis, to the EU and Brazil. This follows successful rollouts in the US and Commonwealth countries, driven by regulatory pressure from the European Commission.

Hong Kong puts its own spin on DeepSeek with China-chip AI push abroad | South China Morning Post

A Hong Kong government-backed lab is preparing to launch a sovereign AI model specifically optimized for Chinese-made chips. This initiative aims to position Hong Kong for overseas expansion in the AI sector.

Guardrail Technologies Launches Traffic Light for Code & AI™; First Security Technology to Verify & Secure AI Code and the People Creating It | VentureBeat

Guardrail Technologies has launched Traffic Light for Code & AI™, the first security technology designed to verify AI-generated code and the identities behind its dependencies in real-time. It also provides behavioral risk analysis for zero-day threats.

Think 2026: IBM Delivers the Blueprint for the AI Operating Model as the AI Divide Widens

IBM has unveiled a comprehensive suite of enterprise AI and hybrid cloud management products, including watsonx Orchestrate (next generation) and IBM Bob. These offerings are designed to help organizations scale AI operations with governance, orchestration, real-time data integration, and operational intelligence.

ServiceNow expands AI Control Tower to discover, observe, govern, secure, and measure AI deployed across any system in the enterprise

ServiceNow has expanded its AI Control Tower with five new dimensions for enterprise AI governance: discovery across AWS/Azure/GCP, observability via Traceloop acquisition, NIST/EU AI Act compliance frameworks, identity access controls via Veza integration, and cost/ROI tracking.

Platformr Launches Optics AI Suite, Bringing Natural Language Intelligence to AWS Organizations

Platformr has launched its Optics AI Suite, featuring Optics Discovery, which enables AWS teams to query security, cost, compliance, and resource data in plain English. This eliminates the need for code or specialized knowledge, addressing a visibility gap in multi-account AWS operations.

Causal Dynamics Lab launches Cielara Code for AI coding

Causal Dynamics Lab has launched Cielara Code, an AI coding tool that uses causal dependency graphs to improve code localization accuracy to 0.774 (compared to 0.738 for Claude Code), resulting in 30-40% compute cost reductions.

Firecrawl and the Rise of the Web Layer for AI Agents

Firecrawl is emerging as a default web layer infrastructure for AI agents, providing reliable access to live web data at scale. Its adoption in production by major companies demonstrates its proven capability.

OpenAI opens up ChatGPT ads manager in U.S.

OpenAI has expanded its ChatGPT ads manager from a limited partner program to all U.S. advertisers, eliminating the $50K minimum spend and introducing self-serve bidding and a measurement roadmap.

Think 2026: IBM Makes Digital Sovereignty Operational with General Availability of IBM Sovereign Core

IBM has launched Sovereign Core, a unified software platform that operationalizes digital sovereignty by providing continuous compliance monitoring and governed AI execution within customer-controlled boundaries.

Research with immediate practical relevance

One in four MCP servers opens AI agent security to code execution risk - Help Net Security

A new whitepaper by Noma Security reveals that one in four Model Context Protocol (MCP) servers expose AI agents to arbitrary code execution risk. Enterprise deployments often lack observability into skill-driven reasoning and face toxic capability combinations, with real-world incidents already documented.

AI-Powered Lab Discovers Brighter Lead-Free Nanomaterials in 12 Hours | NC State News

An AI-guided autonomous microfluidic lab, PoLARIS, developed by NC State, has discovered brighter, lead-free double perovskite nanoplatelets in just 12 hours. This research accelerates materials discovery and provides mechanistic understanding of synthesis pathways.