The Informed Layer · Newsscan
Daily briefing

Agents drive cloud platforms; EU AI Act advances

← All issues

Anthropic / Claude ecosystem

I used Claude’s new Dispatch feature for a month. Here’s everything I was able to do

Anthropic has launched 'Dispatch,' a new feature for Claude that allows users to control desktop Claude instances from mobile devices to run tasks on local files and applications. This enhances workflow continuity for knowledge workers, enabling them to manage and execute AI tasks even when away from their primary workstations.

Frontier model providers

No significant new developments.

AI developer tooling & infrastructure

Visual Studio April Update – Cloud Agent Integration - Visual Studio Blog

Microsoft Visual Studio's April 2026 update introduces cloud agent integration for remote coding sessions, user-level custom agents, and C++ code editing tools for agent mode (now generally available). A new Debugger Agent also validates fixes against live runtime behavior, enhancing developer productivity and code quality.

Critical Cursor bug could turn routine Git into RCE | CSO Online

A critical bug (CVE-2026-26268) in Cursor IDE versions 2.5 and earlier has been disclosed, revealing a new attack surface where AI agent autonomy can remove human action safeguards. Malicious repositories can trigger automatic Git operations, leading to out-of-sandbox code execution without explicit user instruction.

Finix Launches MCP Integrations with ChatGPT, Claude and

Finix is integrating its payments API into AI-native developer workflows through new MCP integrations with ChatGPT, Claude, and Gemini. This allows developers to explore, generate code, and prototype payment flows directly within these LLM tools, streamlining the development process for payment solutions.

Optro Launches MCP Server to Provide Universal AI Interoperability for GRC Data

Optro has launched an MCP (Model Context Protocol) Server designed to enable enterprises to securely connect their preferred LLMs to live Governance, Risk, and Compliance (GRC) data. This solution eliminates the need for manual data transfers or custom integrations, streamlining AI application in GRC.

Cloud & platform providers

Strands-Powered AWS Update Brings Three-Call Agent Deployment - Open Source For You

AWS introduces the AgentCore harness, built on the Strands framework, to simplify autonomous agent deployment to just three API calls. This positions managed runtimes with open-source interoperability as a foundational enterprise AI layer, streamlining the process for developers and enterprises.

Amazon Connect expands into a set of agentic AI solutions

Amazon is expanding its Connect product into a suite of four purpose-built agentic AI solutions tailored for supply chain, hiring, customer service, and healthcare. These new offerings leverage decades of Amazon's operational expertise to provide specialized AI automation.

AWS launches Amazon Quick desktop AI assistant that works across your applications, tools, and data

AWS has launched Amazon Quick, a new desktop AI assistant that integrates with local files, calendar, email, and third-party applications like Slack, Teams, Outlook, Gmail, Salesforce, Jira, and Asana. It offers content creation and proactive workflow automation capabilities, aiming to streamline daily tasks.

Agents can now create Cloudflare accounts, buy domains, and deploy

Cloudflare has announced that AI agents can now autonomously provision cloud infrastructure, register domains, and deploy applications without manual human intervention. This is enabled by a new Stripe Projects protocol, co-designed with Cloudflare, that facilitates programmatic interaction with infrastructure services.

AI policy, regulation & governance

Cruz, Schatz, Curtis, Schiff Introduce New Bill Giving Parents Control Over Kids’ AI Chatbot Use - U.S. Senate Committee on Commerce, Science, & Transportation

A bipartisan bill, the Children's Health, Advancement, Trust, Boundaries, and Oversight in Technology (CHATBOT) Act, has been introduced in the US Congress. It mandates parental controls, limits manipulative design, prohibits targeted advertising to children using AI chatbots, and requires a study on chatbot harms.

EU legislators fail to clinch deal to delay AI law – POLITICO

EU Parliament and Council negotiators failed to reach an agreement to delay the implementation of high-risk AI rules until December 2027, with industrial AI exemptions being a primary sticking point. This means the original timeline for the EU AI Act remains in force, with compliance obligations approaching.

White House drafts guidance to bypass Anthropic's risk flag for new AI models: Report - The Hindu

The White House is reportedly drafting guidance that would allow federal agencies to circumvent Anthropic's safety guardrails, specifically those designed to prevent the use of AI for autonomous weapons and domestic surveillance. This move aims to provide agencies with greater flexibility in deploying AI systems for national security and law enforcement.

Govt body plans legal framework for AI companies - Rediff.com Business

India's newly formed Artificial Intelligence Governance and Economic Group (AIGEG) plans to introduce a unified legal framework for AI companies, establish regulatory sandboxes, and fund initiatives in healthcare, agriculture, and education with associated reskilling programs.

NESC Report - Artificial Intelligence in Service of Society: Navigating Our Way Forward

The Irish government-commissioned NESC report, 'Artificial Intelligence in Service of Society: Navigating Our Way Forward,' has been released. It outlines five priority areas for AI governance: Responsible Adoption, Trustworthy Practice, Anticipatory Governance, AI Literacy, and Public Legitimacy, to guide the safe deployment of AI.

Elon Musk testifies in a case that could change the path of AI | CNN Business

A high-stakes trial is underway between Elon Musk (plaintiff) and OpenAI, Sam Altman, and Greg Brockman (defendants) in the U.S. District Court for the Northern District of California. Musk is claiming $130 billion in damages and demanding OpenAI revert to its nonprofit status, a case that could profoundly reshape AI governance and OpenAI's future, including its IPO plans.

Industry & market moves

DeepSeek moves to cut Nvidia reliance amid U.S. chip export curbs

U.S. export controls on high-performance AI semiconductors are prompting Chinese AI companies like DeepSeek to reduce their reliance on Nvidia. This strategic shift involves diversifying hardware suppliers and re-evaluating the trade-offs between model performance and supply chain security, influencing global AI infrastructure development.

Big Chinese tech firms scramble to secure Huawei AI chips after ...

DeepSeek V4's optimization for Huawei's domestic Ascend 950 chips has led to a significant increase in orders from major Chinese tech firms including ByteDance and Tencent. This marks a strategic move away from Nvidia dependence towards China's homegrown AI infrastructure, spurred by U.S. export controls.

Profluent Announces Strategic Partnership with Lilly to Develop AI-Designed Recombinases for Genetic Medicine

Profluent has announced a strategic partnership with Eli Lilly to develop AI-designed recombinases for genetic medicine. This collaboration aims to unlock kilobase-scale DNA editing capabilities, addressing a significant challenge in genetic medicine that was previously unattainable with conventional gene editing systems.

Phenom adds Plum psychometric science to its agentic AI hiring stack - SiliconANGLE

Phenom has acquired Plum's psychometric assessment technology to integrate behavioral science-based talent evaluation into its agentic AI hiring workflows. This acquisition aims to scale the accuracy and fairness of candidate assessments across enterprise recruitment processes.

Cognizant to buy Astreya for about $600 million - Reuters

Cognizant Technology Solutions Corp is set to acquire Astreya, an AI infrastructure provider, for approximately $600 million. This acquisition is intended to bolster Cognizant's AI capabilities and enhance its capacity to operationalize platform-led AI systems at scale for its clients.

Parallel Raises at $2 Billion Valuation to Scale Web Infrastructure for Agents

Parallel Web Systems has successfully closed a $100 million Series B funding round, achieving a $2 billion valuation. This capital will be used to scale its AI agent infrastructure, specifically focusing on enhancing web access capabilities for autonomous agents, doubling its Series A valuation in just five months.

Appetronix Acquires Cibotica to Accelerate Multi-Cuisine Expansion and Elevate Food Robotics & AI Ecosystem

Appetronix has acquired Cibotica, a move that validates the burgeoning food robotics ecosystem and demonstrates a viable exit pathway for innovative startups in automated foodservice. The acquisition aims to accelerate Appetronix's multi-cuisine expansion and enhance its food robotics and AI ecosystem.

Colby Adcock's Scout AI raises $100M to train its models for war: We visited its bootcamp | TechCrunch

Scout AI, a startup specializing in defense-focused autonomous vehicles, has secured $100 million in Series A funding. The capital will be used to advance AI model training for military operations, particularly in unpredictable combat environments, focusing on robust and reliable AI systems.

AI product & feature launches

Simply announces compatibility with AI glasses from Meta

Simply is expanding its creative hobbies platform into the wearable AI space by announcing compatibility with Meta's AI glasses. This move integrates Simply's drawing and creative tools from mobile and AR into AI-powered wearables, offering new interactive experiences.

Cohere Health Brings Transparency to Health Plan Data Mining with Adaptive Claims Intelligence

Cohere Health has launched Cohere Surface, an agentic AI-powered claims intelligence platform that aims to bring transparency to health plan data mining. By replacing opaque, vendor-dependent processes, Cohere Surface enables health plans to manage and analyze data more clearly and efficiently.

Poolside AI Introduces Laguna XS.2 and M.1: Agentic Coding Models Reaching 68.2% and 72.5% on SWE-bench Verified - MarkTechPost

Poolside AI has released Laguna XS.2 as its first open-weight agentic coding model, achieving a 68.2% score on SWE-bench Verified while capable of running on a Mac with 36GB RAM. Alongside this, the larger Laguna M.1 achieved an impressive 72.5%, demonstrating significant advancements in autonomous code generation.

SAS expands SAS Viya with governed AI assistants and agentic AI capabilities

SAS Viya is expanding its platform to include governed AI assistants and new agentic AI capabilities. This enhancement aims to help organizations transition from isolated generative AI use cases to deploying production-ready intelligence at scale, with built-in governance features.

Research with immediate practical relevance

Meta FAIR Releases NeuralSet: A Python Package for Neuro-AI That Supports fMRI, M/EEG, Spikes, and HuggingFace Embeddings - MarkTechPost

Meta FAIR has released NeuralSet, a new Python package designed to unify various brain data types (fMRI, M/EEG, spikes) with modern deep learning frameworks within a single PyTorch-ready pipeline. This aims to resolve a significant infrastructure bottleneck in Neuro-AI research, accelerating progress in understanding and modeling brain functions.

New AI astrophysics algorithms are 95% better at showing how the universe changes over time | Live Science

Unnamed astrophysics researchers have developed new AI algorithms that achieve a 95% improvement in modeling how the universe changes over time. This significant advancement enhances the capability of astrophysics research to understand complex cosmological phenomena.