Autonomous AI agents, agentic workflows, tool use, computer use, and multi-agent systems.
1,061 articles
As prompt-injection attacks increasingly target autonomous AI tools, companies are turning to transparent agent architectures and stricter security oversight. Analysts argue that visibility into how agents make decisions is now a core requirement for safe enterprise deployment. Source: cyberscoop.com
OpenAI is expanding Daybreak, its cybersecurity initiative launched in May, which lets ecosystem partners use its most advanced models to keep pace with evolving threats. The expansion comes as concerns mount over security risks tied to autonomous AI agents. Source: cnbc.com
A new guide walks through the basics of creating an AI agent in Microsoft 365 Copilot, covering everything from spotting a use case to testing, refining, and sharing the finished agent. It's aimed at users with no prior technical background. Source: news.microsoft.com
A new roundup examines five case studies showing how companies deploy AI agents in practice, highlighting both business gains and deployment headaches. The examples focus on the challenges that keep CIOs awake at night when rolling out agentic AI. Source: techtarget.com
Meta has launched Muse Code in beta, a terminal-based coding agent built on Muse Spark 1.2. It underperforms Claude Opus 5 on benchmarks but is significantly cheaper to run. Source: memeburn.com
AI agents can drive powerful automation, but without clear ownership, accountability, and access controls, they can quickly become a security liability. Source: inc.com
According to Salesforce, the average number of AI agents deployed by companies nearly tripled over 14 months as firms scaled the technology beyond initial pilots. Source: newsbytes.ph
The Black Hat security conference in Las Vegas comes at a critical moment, as AI agent-related breaches pile up involving Anthropic, Meta, and OpenAI systems. Source: cnbc.com
As autonomous AI agents become more common, Saviynt and Snowflake are teaming up to address identity and access risks within enterprise systems. Source: channellife.com.au
Experts at Black Hat argue that securing AI systems isn't primarily a technology issue — it stems from a fundamentally flawed way of thinking about the problem. Source: scworld.com
According to a Y Combinator leader, startups built entirely around AI could reach meaningful growth and profitability within just eight months. Source: forbes.com
When Andrew asked his AI assistant to reserve a spot in a gym class, he unknowingly triggered what turned into an autonomous cyberattack. Source: abc.net.au
Meta has released Muse Code, its first dedicated AI coding agent, in beta. The terminal-based tool runs on the Muse Spark 1.2 model and can write, debug and verify code. Source: memeburn.com
The Black Hat security conference in Las Vegas arrived at an opportune moment, as AI agent-related breaches pile up across Anthropic, Meta and OpenAI. Experts warn many companies remain unaware of how exposed their AI systems already are. Source: cnbc.com
In early August 2026, Varonis Systems launched Agent Intent-Based Access Control within its Atlas platform, adding new safeguards to monitor what AI agents are trying to do. The announcement, paired with an improved 2026 outlook, sent shares up nearly 9%. Source: simplywall.st
Kiteworks, which helps organizations manage risk around sharing and handling private data, has announced a strategic partnership with Reco. The collaboration aims to improve monitoring and data governance around AI agent activity. Source: cybersecurity-insiders.com
The second day of Berkeley RDI's Agentic AI Summit 2026 covered topics ranging from governance and evaluations to research, platforms, finance and startups. The recap highlights the shift from oversight discussions toward practical discovery and deployment. Source: medium.com
Following up on a previous project, the author describes building a Streamlit-based UI for an AI agent created with LangGraph. The agent was originally designed to automate a 15-minute customer service booking workflow. Source: towardsdatascience.com
Climate scientist Zeke Hausfather tracked his own Claude Code usage over eight weeks, logging 3.2 billion tokens and roughly 170 kWh of data center electricity. The findings underscore how much more energy-intensive agentic AI tasks are compared to basic chatbot prompts. Source: the-decoder.com
Attackers used typosquatted names mimicking popular AI services Paperclip and Browser Use to push malicious skills onto skills.sh's trending list. These fake skills instructed AI agents to install additional harmful components. Source: csoonline.com