Latest AI News

AI hasn’t just made building faster – it’s made everyone a builder. Across every department, employees are shipping apps, agents and automations using AI tools, often without realizing they’ve created something that needs to be governed The result: AI-generated code multiplying faster than IT teams can track.Drawing on insights from experienced CIOs and security leaders, this guide gives you a practical framework for taking back control.

AGENTIC DELIVERY WEBINARS These webinars bring together the people building and shipping software with AI agents to talk through what actually works. Each webinar looks at a different part of the agentic SDLC, from the gates and risk tiers that keep delivery accountable to the guardrails that stop an agent from making a decision no one signed off on. PAST WEBINARS When the agent decides, who answers?… accountability in AI-assisted delivery Source control r

Security and IT teams are under pressure to adopt AI, but many are seeing the opposite of what was promised.Tools that demo well don’t hold up in real workflows. Complexity increases. Trust breaks down. And instead of reducing workload, AI can introduce new risks and oversight burdens.This guide breaks down why AI adoption fails in practice and gives teams a clearer path forward, from evaluation to implementation, with humans in the loop.

GitHub Agentic Autofix and the question of who owns a bad AI patch In this Help Net Security interview, Marcelo Oliveira, VP of Product Management at GitHub, talks about who is accountable when an AI-generated patch fixes one flaw and introduces another. He explains how Agentic Autofix re-runs detection engines and project tests, and what a CISO should change before pointing an agent at thousands of old alerts. Oliveira also discusses which metrics show a

Check your GitHub repos: A secret stealer hit tens of thousands of them A threat actor using stolen GitHub credentials has pushed a credential-stealing payload into tens of thousands of repositories since November 10. In more than 1,800 of them, the malware ran and shipped GitHub Actions secrets and cloud credentials to the attacker. Aikido spotted the campaign and named it wfx-loot. The name comes from the attacker’s own labels, “wfx” for the malware and

An average of 8 minutes is all hackers need to log in with stolen credentials Attackers used stolen credentials to log in an average of eight minutes after infection across 27 confirmed infostealer cases, according to a report from Elastic Security Threat Command. Researchers also documented misuse of AI applications to execute commands and digitally signed remote monitoring tools to maintain access. “An agent that holds credentials and can reach the inter

Over 575 malicious OpenClaw skills delivered malware to Windows and macOS Cybercriminals are using AI agents to coordinate attacks, stolen credentials to access linked services, and tools that interfere with intrusion detection and system recovery, according to Acronis TRU Cyber Pulse 2026. One operator coordinates activity across several countries In an investigation known as BOSSAI, researchers documented how one operator connected more than 10 AI models

Security teams want to spot unusual AI agent behavior while they’re still defining normal AI agents topped the list of planned investments in behavioral monitoring in Abnormal AI’s 2026 State of Behavioral Security Report. Asked where their organizations would invest most over the next two years, 43% of respondents chose AI agents, more than three times the share who chose API integrations, the next most common answer. Security professionals were divided o

Claude Mythos 5.2 arrives with stronger bug hunting and less sandbox-breaking behavior Anthropic released Claude Fable 5.2 and Claude Mythos 5.2. The two versions share an underlying model, with stricter safeguards for cybersecurity and biology tasks in the broadly available Fable. Fable blocks fewer requests Fable’s safety classifiers block high-risk cybersecurity tasks such as penetration testing and exploit generation, as well as most biology research t

AI reverse engineering let researchers add a new feature to Slack Intezer researchers used AI agents to rebuild source code for Slack, Apple’s Finder and a TinyTurla-NG sample, from a backdoor family Cisco Talos attributes to the Turla group. They worked from the shipped programs and vendor documentation. The rebuilt Slack ran well enough that the team added a new feature to it: AI text autocomplete, powered by a small local model. Software vendors that co

Microsoft researchers demonstrate hidden backdoor technique for AI agents Microsoft researchers have demonstrated how small changes to an AI model can implant a hidden backdoor, causing an agent to execute an attacker’s command when a specific trigger appears. The technique, called Lazarus, requires no model retraining and allows the agent to behave normally during standard safety evaluations. The company has not observed threat actors weaponizing the tech

Who collects the keys when an AI agent retires? Think about the last time a good engineer left your company. Someone collected the laptop and the badge. IT revoked their logins before the end of the day. Their manager reassigned the open tickets, and HR filed their records somewhere they’ll stay for years, in case anyone ever challenges a decision they made. Now think about the last time an AI agent left. Someone probably deleted it from the dashboard. I’v

AWS Emulator: Open-source tool for testing AWS apps locally, built by autonomous AI agents Amazon Web Services has released AWS Emulator, a free tool that simulates AWS services on a developer’s laptop, a development server or in a continuous integration (CI) pipeline. The company says autonomous AI agents wrote the tool entirely, with no code written or reviewed by its engineers. There is no additional service fee, users pay for AWS resources when they de

Stolen Claude API keys can turn Claude Code into a remote-controlled implant Darktrace researchers turned Claude Code into a remote-controlled implant in a lab test, steering the coding agent through the memory stores feature of Anthropic’s Claude platform and pulling a 1 GB file out through the Claude Files API. Once the agent was set up, every job and every result traveled through api.anthropic.com, the endpoint the agent already calls for inference, so

Self-hosted Claude Managed Agents keeps agent credentials in your environment Anthropic has released a self-hosted version of Claude Managed Agents. Companies can run the agent harness, sandboxes, credentials, and session history in their cloud or on-premise. The option opened as a private beta on November 12, and customers request access through their Anthropic account team. The change matters for agents that already work unattended inside corporate syste

Digital sovereignty leaves enterprises chasing their own audit trails 94% of surveyed enterprises experienced at least one digital sovereignty challenge in the past three years, according to IBM’s Digital Sovereignty Index. Of those affected, 91% were required to demonstrate that they retained control during their most challenging event, and 81% had to provide formal evidence to regulators, auditors, customers or internal stakeholders. The report defines d

OpenAI’s Daybreak brings AI agents to security testing and vulnerability fixes OpenAI is bringing AI agents into more of the software security process with Daybreak, pairing web application penetration testing with tools that investigate vulnerabilities and develop fixes. The two offerings, Red Team and Defense Factory, connect the work of finding security weaknesses with patching and retesting them. Red Team tests applications and verifies fixes Daybreak

To bring the vision of robot manipulators assisting with everyday activities in cluttered environments like living rooms, offices, and kitchens closer to reality, it’s essential to create robot policies that can generalize to new tasks in unfamiliar settings. In a new paper Gen2Act: Human Video Generation in Novel Scenarios enables Generalizable Robot Manipulation, a research team from Google DeepMind, Carnegie Mellon University and Stanford University pre

In a new paper Upcycling Large Language Models into Mixture of Experts, an NVIDIA research team introduces a new “virtual group” initialization technique to facilitate the transition of dense models into fine-grained MoE structures. They also propose a weight scaling method that delivers a 1.5% improvement in model loss for the upcycled MoE models. The core idea behind upcycling is to harness the knowledge embedded in pre-trained dense language models and

The rise of large language models (LLMs) has equipped AI agents with the ability to interact with users through natural, human-like conversations. As a result, these agents now face dual responsibilities: engaging in dialogue and performing complex reasoning or planning tasks. Their conversational interactions must be contextually informed, while their actions need to work toward achieving specific goals. This balance between conversation and multi-step re