CISA added CVE-2026-55255, an access-control flaw in the Langflow visual framework for building AI agents, to its Known Exploited Vulnerabilities catalog. The issue is an insecure direct object reference in the /api/v1/responses endpoint that allowed one authenticated user to invoke another user's flows. Attackers have already abused it to steal AI and cloud credentials from affected deployments. Langflow version 1.9.2 or later contains the patch.

Source: AI Agent Store News

Akeneo announced Agentic Ziggy, an agentic orchestration layer inside the Akeneo Product Cloud that coordinates specialist agents for data modeling, schema mapping, enrichment, and continuous quality checks (announced July 8, 2026).

Source: AI Agent Store News

Peraton announced Peraton[x], an enterprise agentic AI platform built for government operations and other high-stakes missions. The company says the system is meant to boost efficiency without replacing people, and it launched from Reston, Virginia on July 7, 2026.

Source: Agentic.ai News

AIAI Holdings said its portfolio company Constellation Network publicly launched Gate AI on July 8, 2026. The security layer is built to block prompt injection, reduce token costs, and verify audit trails through Constellation Digital Evidence.

Source: Agentic.ai News

M-Files launched M-Files for Consulting on July 9, 2026, a purpose-built app for consulting firms. The product organizes work from proposal to delivery and adds governed access to confidential client information.

Source: Agentic.ai News

OpenAI launched its GPT-5.6 family featuring Sol, Terra, and Luna models on July 9, 2026, with a strategic desktop consolidation. The flagship Sol model nearly matches Anthropic's Claude Fable 5 on benchmarks.

Source: Forbes

GPT-5.6 cleared government review and became publicly available on July 9, 2026, after initially being released to 20 vetted organizations on June 26, 2026. The Sol model achieved 88.8% on TerminalBench.

Source: Tech Insider

Alongside the GPT-5.6 public launch on July 9, OpenAI launched ChatGPT Work, an agentic productivity tool combining Codex technology with third-party integrations to operate across apps and complete multi-step projects.

Source: BuildFastWithAI

CVE-2025-47981 is a heap-based buffer overflow in Windows SPNEGO with CVSS 9.8, network attack vector, requiring no credentials or user interaction, and is fully wormable. SPNEGO is enabled by default on every Windows version from 10 1607 and Windows Server 2008R2 onward.

Source: ByteIOTA

RoguePlanet (CVE-2026-50656) is a race condition privilege escalation vulnerability where POC code posted on GitHub can result in System privilege shell access on compromised systems. Microsoft issued an emergency patch on July 9.

Source: Help Net Security

Microsoft released July 2026 security updates addressing 622 Microsoft-assigned CVEs and 428 Chromium vulnerabilities, with organizations prioritizing critical flaws under active exploitation including BitLocker security bypass and Active Directory Federation Services privilege escalation.

Source: NT Compatible

FortiBleed is a sustained campaign where compromised Fortinet firewalls are being used to deploy ransomware, with 74,000 stolen credentials circulating and at least 12 confirmed ransomware infections.

Source: Threat-Modeling.com

An unauthenticated attacker with network access via HTTP can completely compromise Oracle Payments through CVE-2026-46817. Threat actors have been observed exploiting this vulnerability on Oracle E-Business honeypots, and Oracle addressed it in its July 2026 Critical Patch Update.

Source: Threat-Modeling.com

Released on July 8, 2026, VS Code 1.128 brings richer multi-chat agent sessions, generally available image support in Chat, and OS-level keyboard shortcuts. Copilot Vision now allows users to attach images and PDFs to Chat by pasting, dragging, or dropping them.

Source: Visual Studio Code Official

This changelog covers VS Code v1.123 through v1.127, shipped throughout June and early July 2026. Agentic browser tools are now generally available, allowing agents to navigate pages, inspect content, capture screenshots, and validate web apps directly in VS Code.

Source: GitHub Changelog

AWS Builder Center celebrated its one-year anniversary on July 9, 2026, with 5,548 authors publishing 6,448 articles. EKS Auto Mode and ECS Managed Instances reduced GPU management fees by up to 60% starting July 1, 2026. Amazon Aurora DSQL CDC is now generally available.

Source: AWS News Blog

AWS announced new features on July 8, 2026, including Amazon Connect Tasks and Emails workforce planning enhancements and additional service updates covered in the weekly announcement video.

Source: AWS (YouTube)

Services moving to maintenance will no longer be accessible to new customers starting July 30, 2026, including Amazon Bedrock Agents Classic and AWS IoT Device Defender features. Updates published June 30, 2026.

Source: AWS News Blog

AWS announced a $1 billion investment in Forward Deployed Engineering (FDE) embedding thousands of AI engineers with customers, and CLEAR integration with Amazon Connect for secure identity verification on July 1, 2026.

Source: CXToday

Anthropic launched Claude for Teachers, providing verified K-12 educators in the US with free access to premium Claude capabilities, teaching skills, and curriculum resources aligned to academic standards in all 50 states. The program includes new education connectors, open-source teaching skills, and a pilot program with Detroit Public Schools Community District.

Source: Releasebot

Anthropic hired Tom Blomfield, co-founder and former CEO of British fintech Monzo, who left Y Combinator to join the company's AI compute team on July 13, 2026. This continues Anthropic's aggressive talent acquisition strategy following additions like Google DeepMind's John Jumper and former Tesla AI leader Andrej Karpathy.

Source: Tech Startups

Anthropic is in contact with Samsung to explore collaboration on a custom AI chip, continuing plans first reported in April 2026. The company hasn't yet decided the chip's specifications or use cases, but the move reflects Anthropic's strategy to diversify its hardware stack beyond Nvidia, Google, and Amazon chips.

Source: TechCrunch

California Governor Gavin Newsom announced on June 29, 2026 a historic statewide AI deployment with Anthropic, offering all state agencies and local governments access to Claude at 50% discount. The deal includes workforce training and technical assistance, with the AI assistant 'Poppy' on track for full statewide rollout in July 2026.

Source: Build Fast with AI

Anthropic published research on mechanistic interpretability discovering a 'J-space' within LLMs that could reveal hidden model behaviors not apparent in outputs. The research, published July 13, 2026, represents Anthropic's ongoing work on understanding AI model internals for safety monitoring.

Source: MIT Technology Review