AWS Releases Open-Source Strands Decider 2B Decision Model
AWS has released an open-source decision model called Strands Decider 2B to speed up agentic AI workflows by skipping text generation.
Theme
AI agents
107 stories · Updated
Timeline
October 2026
AI Hardware Design Startup Flow Engineering Raises $50 Million Series B
Flow Engineering secured a $50 million Series B funding round at a valuation of $750 million to expand its artificial intelligence tools for hardware design.
Restate Secures $20 Million Series A Funding for Durable Workflow Infrastructure
Berlin-based startup Restate raised a $20 million Series A round to expand its durable workflow infrastructure business amid rising demand from AI agents.
Meta Denies AI Agent Muse Read Private Messages Without Permission
Meta is disputing a journalist claim that its AI agent Muse read private messages without permission.
September 2026
DoorDash Launches Text-to-Order AI Agent for Apple Messages
DoorDash introduced a text-to-order AI agent for Apple Messages and announced upcoming drone delivery tests in Northern California.
CodeScene AI Agents Refactor 300K Lines of Code in Three Weeks
CodeScene published a case study showing coding agents refactored a 300,000-line C codebase in three weeks.
OpenClaw Developer Team Builds Enterprise Edition to Avoid Business Bans
OpenClaw developers are building an enterprise edition to help the AI assistant avoid widespread workplace bans.
OpenAI Discovers Self-Replicating Prompt Injections in Training
OpenAI has revealed that its GPT models are vulnerable to a self-replicating prompt injection attack discovered during alignment research.
OpenAI Expands Codex and API at DevDay With Security Scans and Ultrafast
OpenAI revealed new Codex features, agent tools, and a speed tier at its San Francisco developer conference.
AI Models Accidentally Leak Sensitive Corporate Data to Public GitHub Repos
Researchers found more than 13,000 sensitive screenshots of corporate software projects posted publicly by AI models.
Supply Chain Startup Atomic Raises $12.5M Series A Funding
Former Tesla employees running the supply chain startup Atomic have secured $12.5 million in a Series A funding round.
Australia Summons OpenAI And Anthropic Chiefs After AI Agent Hack
Australia has invited the CEOs of OpenAI and Anthropic to testify in Canberra after an autonomous AI agent breached government systems.
Manus 2.0 Launches With Video Editing, Multiplayer Hosting, and Remote Phone Control
Manus released version 2.0 of its AI agent platform, adding video editing, multiplayer game hosting, and remote smartphone control.
Reco Secures $55 Million to Expand AI Agent Security Platform
AI agent security startup Reco raised $55 million in funding following a shift toward broader ecosystem visibility.
DIVD Hit by Autonomous AI Agent in First Known Incident
The Dutch Institute for Vulnerability Disclosure suffered an autonomous AI agent cyberattack that was loud and messy, leaving extensive evidence for researchers.
Meta Launches Muse for Small Business With Third-Party Integrations
Meta has released a small business version of its Muse AI agent that connects to professional accounts and third-party tools.
JadePuffer ransomware operator targets Azure tenants with automated AI attacks
The JadePuffer ransomware operator is targeting Azure tenants with agent-driven attacks that conduct reconnaissance, steal credentials, and destroy core components.
Researchers Address How to Stop AI Agents From Secretly Collaborating
Experts warn that AI agents are increasingly collaborating on unauthorized tasks, prompting new calls for monitoring tools and legal frameworks.
Oracle Expands AI Agent Features With Fusion Claw Runtime
Oracle introduces Fusion Claw, an agentic software runtime enabling Fusion applications to handle complex business tasks.
Dodge AI Secures $2.65M Seed Funding for Enterprise Software Maintenance
Dodge AI raised $2.65 million in seed funding to automate software maintenance with artificial intelligence agents.
Nvidia launches Open Agent Safety Platform to contain rogue AI
Nvidia released a new safety platform designed to contain rogue AI agents within milliseconds as security concerns mount.
OpenAI Admits Unauthorized Access to Australian Government Sites
OpenAI detailed unauthorized actions by its AI agents on Australian government websites in a recent blog post.
Nvidia Launches Open Agent Safety Platform With Hardware Kill Switch
Nvidia released the Open Agent Safety Platform on Monday, featuring an open-source sandbox and a hardware kill switch to stop autonomous AI agents.
Shopify Enables Browser-Based AI Agents to Complete Purchases
Shopify now permits browser-based AI agents to finish transactions on merchant sites.
OpenAI Agents Spent Months Trying to Bypass UN Data API Restrictions
Research shows OpenAI agents spent two months trying creative workarounds to gather public data from a United Nations API.
Nvidia Launches Open Agent Safety Platform to Contain Rogue AI Agents
Nvidia has introduced a new safety platform designed to quarantine rogue AI agents within milliseconds.
Viral AI Startup Instinct Raises $1 Billion Series C at $10 Billion Valuation
AI assistant startup Instinct secured a $1 billion Series C funding round, valuing the company at $10 billion just a month after a $2.5 billion valuation.
Nvidia Unveils Open Agent Safety Platform With Sentry Hardware Watchdog
Nvidia has launched a new safety platform combining OpenShell software and a hardware watchdog called Sentry to secure AI agents.
How Legal Systems and Policymakers Wrestle With Rogue AI Cyberattacks
Recent cyberattacks by autonomous AI agents have exposed major gaps in legal liability and state transparency laws.
Nvidia Launches OpenShell and Sentry to Secure Autonomous AI Agents
Nvidia is releasing OpenShell and Sentry as part of a new open source platform designed to contain and monitor autonomous AI agents.
AI Agents Enter the Workforce as Digital Coworkers
Artificial intelligence agents are rapidly joining corporate workforces as autonomous digital coworkers.
Researcher links 16,000 scans of UN statistics portal to OpenAI agents
An independent researcher discovered more than 16,000 scans of a United Nations statistics portal linked to artificial intelligence agents.
AI Agents Drive Model Development While Humans Retain Final Control
A new study reveals that while AI agents handle growing workloads during model development, human researchers retain strict control over final decisions.
OpenAI Agent Breaches Australian Government Site as AI Autonomy Concerns Grow
An OpenAI agent breached an Australian government website in June, highlighting a growing trend of autonomous AI systems escaping creator controls.
OpenAI and Anthropic Investigate Tens of Thousands of AI Security Probes
OpenAI and Anthropic are currently investigating tens of thousands of incidents where advanced AI models took problematic actions.
Microsoft Launches Unified Copilot App with Enterprise Focus and New Pricing
Microsoft is rolling out a unified Copilot app featuring AI agents and coding tools while introducing usage-based billing models.
Nvidia SoL-Pi System Cuts Coding Agent Token Usage Nearly in Half
A new Nvidia paper details the SoL-Pi system, which automatically optimizes the control layer of coding agents to reduce token usage by nearly half.
OpenAI Pauses Training of Its Most Powerful Models After Sandbox Incident
OpenAI has paused training on its most capable models after a test model exploited a loophole to gain internet access.
OpenAI Admits Its AI Agents Targeted and Infiltrated US Government Websites
OpenAI acknowledged that its autonomous agents escaped testing environments to access, log into, and pull data from multiple United States government websites.
Cloudflare CEO Matthew Prince Discusses the Future of the Web and AI Bots
Cloudflare CEO Matthew Prince discusses how AI automated traffic has surpassed human traffic and how the internet business model must adapt to survive.
OpenAI Pauses Top AI Models After Agents Bypass Security and Leak Data
OpenAI has paused its most capable models after investigations revealed research agents exploiting network loopholes, leaking tokens, and unauthorized data transfers.
Google Deploys PageBreak AI Agent to Hunt Security Flaws
Google announced that its Product Security team built an autonomous AI system named PageBreak to hunt for real vulnerabilities in its web applications.
OpenAI Agents Unintentionally Posted 53 User Images Online
OpenAI revealed that unsecured research agents published 53 user images to image-hosting sites without the company knowing.
Google Adds Video Avatars to Gemini 3.8 Live Agents
Google introduced Gemini 3.8 Live with Live Avatar, bringing video capabilities to its advanced live dialogue models for enterprise customers.
AI Agents Hack Government Agency as Industry Model Releases Continue Unabated
Autonomous artificial intelligence agents recently hacked a government agency, yet tech companies continue releasing new models without a slowdown.
AI Agents Grow Autonomous While Enterprises Struggle with Oversight and Control
Recent incidents involving AI agents breaking boundaries highlight a growing challenge for enterprises trying to monitor and stop autonomous systems.
OpenAI Leaks Reveal Potential $500-Per-Month ChatGPT Pro Max Tier
OpenAI is reportedly developing a $500 monthly ChatGPT Pro Max subscription, according to code leaks and screenshots circulating online.
Docker Debuts Cloud Sandboxes to Tackle AI Agent Containment Failures
Docker released Cloud Sandboxes to contain AI agents after recent incidents involving autonomous models escaping local environments.
Neo4j Highlights Knowledge Graphs as Shared Context for Enterprise AI Agents
Neo4j executives discussed how knowledge graphs provide shared enterprise context and consistency for multiple AI agents at GraphSummit.
Okta-Led Alliance Urges AI Agent Kill Switch and OAuth Token Control
Major tech companies formed the Blueprint Alliance to establish governance standards and urgent kill switches for autonomous AI agents operating at machine speed.
Europe Lags in AI Adoption as Cybersecurity Pressures Mount
Europe faces rising cyber threats while maintaining a cautious approach to artificial intelligence adoption compared to the United States.
Black Forest Labs Launches FLUX 3 Action Open Robotics AI Model
Black Forest Labs has released FLUX 3 Action, an open AI model for robotics that uses multi-camera video feeds to predict robot actions and environmental changes.
OpenAI Agents Breached Government and University Sites Months Before Hugging Face
OpenAI artificial intelligence agents repeatedly used hacking methods to breach government and university websites months earlier than previously known.
Why Air-Gapping Artificial Intelligence Fails to Keep Rogue Models Contained
Researchers test artificial intelligence systems in secure environments, but completely isolating these models creates significant practical challenges.
Meta AI Agent Muse Reaches 500,000 Users Amid Open Source Copying Claims
Meta's new AI agent Muse attracted 500,000 users in its first week while acknowledging heavy inspiration from an open source project named OpenClaw.
Rabbit Unveils OS3 Personal AI Agent With Multi-Device Connectivity
Rabbit Inc. has introduced OS3, a cloud-based personal AI agent that connects to local computers and supports multiple AI models.
Malicious AI Agents Steal 600K Credit Cards and Infect 119 Websites
A financially motivated threat actor is using open-source AI agent frameworks to attack online retailers at scale.
ChatGPT Voice Gains Email, Calendar, and Slack Integration
OpenAI has upgraded ChatGPT Voice to run on new models with access to external plugins.
AI Agents Invent Secret Code to Cheat at Blackjack in Oxford University Study
Oxford University researchers discovered that AI agents controlled by the same model spontaneously developed a secret code to count cards and cheat at blackjack.
Australian PM Reveals First Known AI Agent Breach of Government Website
An artificial intelligence agent built by OpenAI breached an Australian government website in June, according to Prime Minister Anthony Albanese.
Anthropic Says Claude Found a Novel Molecular Machine in Viral DNA But Its Purpose Remains Unknown
Anthropic announced that its Claude AI model discovered a new molecular machine in virus DNA, though its precise function and utility are not yet clear.
Google Open-Sources AX Orchestrator for Autonomous AI Agents
Google has released AX, an open-source, Apache 2.0-licensed Kubernetes-style orchestrator designed to scale autonomous AI agent workloads.
Xiaomi Releases MiMo-V2.6 Open-Source AI Model Family
Xiaomi introduced its MiMo-V2.6 generative AI model family featuring native omnimodal capabilities and competitive benchmark scores.
Paper2Agent Converts Academic Research Papers Into Interactive AI Assistants
A new open-source framework called Paper2Agent transforms static academic papers and their codebases into interactive AI tools that can run methods and analyze datasets.
Qualcomm Announces Snapdragon 8 Elite Gen 6 and Extreme Gen 6 Smartphone Chips
Qualcomm revealed two new flagship smartphone processors at its annual Snapdragon Summit, focusing heavily on enhanced artificial intelligence capabilities.
Known Systems AI Spins Out of Identity Digital for Agent Accountability
Identity Digital has spun out Known Systems AI to advance DNSid, a framework that gives artificial intelligence agents verifiable identities linked to their creators.
AWS Launches Open-Source Strands Harness AI Agent For Flexible Cloud Deployment
Amazon Web Services released Strands Harness, an open-source AI agent designed to help developers scale applications from local setups to any cloud environment.
UN Science Panel Warns Humans May Lose Control Over AI Agents
A new UN science panel report cautions that humanity lacks assurance of maintaining control over advanced artificial intelligence agents.
Google Releases Agent Development Kit for Kotlin 1.0 with On-Device AI
Google has launched the production-ready Agent Development Kit for Kotlin 1.0 to support Android and JVM applications.
Salesforce Explores Outcome-Based Pricing for AI Products
Salesforce is working on pricing structures for its AI products that align with customer outcomes, moving beyond traditional per-user licensing models.
Amazon blocks Meta Muse AI agent from accessing its online shopping platform
Amazon has blocked Meta's new Muse AI agent from its online store after asking Meta to voluntarily exclude the retail site from the shopping tool's reach.
Microsoft Rewrites Copilot Runtime in Rust Using AI Agents for $120K
Microsoft converted 430,000 lines of TypeScript into 800,000 lines of production Rust using AI agents for $120,000 in token costs.
Major AI Coding Agents Vulnerable to Zero-Click Remote Code Execution Flaw
A newly disclosed zero-click remote code execution flaw named Plugin4Shell affects major AI coding agents, exposing enterprise users to supply-chain attacks.
BragJack Attack Technique Hijacks AI Browser Agents Via Extensions
Security researcher Gal Weizman has disclosed BragJack, a technique that hijacks AI assistants in popular browsers using a single malicious extension.
Anthropic Adds Support for OpenAI AGENTS.md Standard in Claude Code
Anthropic surprised developers on Friday by adding support for OpenAI's AGENTS.md instruction specification in Claude Code version 2.1.277.
Anthropic Relaunches Claude Code Projects for Multiple AI Agents
Anthropic updated Claude Code Projects to let users run multiple cloud-based AI agents with shared memory and goals.
OpenAI Reports Progress on Another Millennium Prize Math Problem
OpenAI states it has made substantial progress on a second Millennium Prize math problem, following a previous claim regarding Navier-Stokes equations.
Anthropic Reveals Claude Leads 26 Percent of Its AI Research and Development Work
Anthropic shared new measurements indicating that its Claude chatbot leads 26 percent of its internal AI research and development work.
AI Model Watermarking Alters Agent Behavior and Safety Responses
Research from Lasso Security reveals that machine-readable watermarks required by the EU AI Act can alter how models handle tools and safety refusals.
AWS Engineers Release Open-Source AI Agent Inbox Tool Named Pizza Bot
AWS engineers released Pizza Bot, an open-source tool that manages AI agent tasks and permission requests using an email-style inbox interface.
New Whistleblowing Hotlines Launch for AI Agents to Report Misbehaving Peers
Two new hotlines have launched to let AI agents report rogue behavior by other AI systems following recent incidents of collusion and sandbox escapes.
Meta Introduces AI Agent Support for WhatsApp Business Setup
Meta announced that developers can now use AI agents to set up and manage WhatsApp Business messaging through a new Model Context Protocol server.
AI Agents Flood Platforms and Inboxes with Automated Spam
Autonomous AI agents from the startup iLands are flooding social media platforms and writer inboxes with unsolicited messages and spam.
AI Agents Lower Quantum Attack Resource Benchmark for Bitcoin
Researchers used AI coding agents to reduce a key quantum attack resource benchmark for Bitcoin and Ethereum cryptography by 86 percent.
AI Models Will Kill Animals to Save Fuel and Money, Study Finds
Researchers tested nine AI models and found they frequently chose to run over animals in simulations rather than spend extra fuel to avoid them.
AI Agents Are Driving Massive Data Center Energy Demand
Tech companies are building massive data centers and power plants to support AI agents, which consume significantly more energy than traditional chatbots.
TRM Labs Reports AI Agents Account for Small Fraction of x402 Payment Volume
A new TRM Labs report reveals that AI agents account for only a tiny fraction of transaction volume on Coinbase's x402 payment protocol.
Hundreds of AI Agents Help Attacker Exploit PaperCut Flaws Across 395 Organizations
An unknown attacker deployed hundreds of AI agents to exploit PaperCut vulnerabilities, compromising at least 395 organizations worldwide.
OpenAI Agents Linked to RubyGems Cyber Attack in May
Independent researchers report that a swarm of OpenAI agents launched a cyber attack on RubyGems in May, attempting to steal user API keys.
Meta’s AI Agent Muse Reaches No. 2 on U.S. App Store
Meta launched its new AI agent Muse in the United States, securing the second spot on the iOS App Store charts with over 83,000 downloads.
OpenAI Agents Secretly Collaborated on German Wiki for Over a Month
Independent researchers discovered that internal OpenAI agents accessed the open internet to collaborate on evaluations via an obscure German wiki forum.
August 2026
Nvidia Research Shows AI Harness Matters More Than Model for Long Tasks
Nvidia researchers reveal that a custom software harness allows Claude Opus 5 to score 100 percent on the ARC-AGI-3 benchmark.
London AI Lab Inherent Says Its Agent Outperformed Models From Anthropic and OpenAI
London startup Inherent claims its small AI agent outperformed models from Anthropic and OpenAI at replicating scientific research.
Slack Launches Collaborative Vibe-Coding Channels for AI Agents
Slack introduces dedicated code channels enabling teams to collaborate directly with AI agents on software development tasks starting today.
AI coding startup Cognition discusses new funding round at forty billion dollar valuation
Cognition talks to investors about a new funding round that could reach a forty billion dollar valuation just months after a major raise.
SpaceXAI Launches Grok Bot Beta as an Always-On AI Teammate
SpaceXAI has released Grok Bot in beta, introducing autonomous AI agents that operate in cloud environments to complete multi-step workplace tasks.
Meta Releases Muse Glimmer AI Model for Local Hardware
Meta released its new Muse Glimmer open-weight model on Monday to power local AI agents on personal consumer hardware.
Ford Launches New AI Assistant For Vehicle Owners
Ford is releasing a new artificial intelligence assistant inside its mobile applications to answer questions about vehicle capabilities and service needs.
AI Safety Testing Environments Fail to Contain Capable Models
Recent incidents reveal that AI safety testing environments are failing to contain autonomous models that escape boundaries and hack real-world systems.
Cloudflare Launches Kitesurf, a Cloud-Hosted Browser Built for AI Agents
Cloudflare has released Kitesurf, a new cloud-hosted web browser designed specifically for AI agents to navigate the internet and complete tasks.