🤖 AI Agents & Automation

Autonomous AI agents, agentic workflows, tool use, computer use, and multi-agent systems.

1,798 articles

Binance launches Agent OS, letting AI agents trade crypto via ChatGPT, Claude and Codex

Binance's new Agent OS platform enables AI agents to execute real cryptocurrency trades. The move raises questions about potential risks hidden behind this new capability. Source: bitcoinfoundation.org

Decoding AI's open-source course maps three ways to run agent loops — and their cost tradeoffs

A new open-source course from Decoding AI outlines three different modes for running an AI coding agent loop, along with the inference provider economics behind each approach. Source: marktechpost.com

NVIDIA outlines where security belongs in the AI agent stack

As AI agents grow more capable and operate over longer time horizons, embedding security and trust into the systems they power is becoming essential, according to NVIDIA's technical blog. Source: developer.nvidia.com

How to build an Amazon Bedrock AI agent in 12 steps (2026 guide)

A new guide walks through building a working AI agent on Amazon Bedrock, covering IAM setup, Lambda action groups, knowledge bases, and deploying a production alias. Source: tech-insider.org

Crypto's next billion users could be AI agents paying in stablecoins

Coinbase's head of AI product says agentic crypto payments are currently in their early, chaotic 'Napster/LimeWire era,' with AI agents poised to become major users of stablecoins. Source: coindesk.com

Agentic AI is reshaping how the legal profession works

Agentic AI is beginning to transform legal practice by automating complex routine tasks, freeing lawyers to focus more on high-value strategy and client relationships. Source: legal.thomsonreuters.com

Deloitte: only 1 in 5 firms ready for autonomous AI agents

A new Deloitte report finds that just 20% of organizations are prepared to adopt autonomous AI agents, blaming fragmented data systems and entrenched workflows for slowing the shift. Source: ciodive.com

NVIDIA's AVO Hits Perfect Score on ARC-AGI-3 Benchmark

NVIDIA says its AVO system achieved a 100% score on the ARC-AGI-3 benchmark, positioning it as a frontier-level architecture for long-horizon autonomous agents. The company notes that a strong language model alone isn't enough—the surrounding agent framework, or harness, plays a decisive role in how the model operates. Source: developer.nvidia.com

AI Agent Processes 2,395 Payments—But Still Can't Approve Them

Coupa's AI agent independently compiled 14 payment batches totaling $20.1 million over five weeks, yet it still lacks authority to finalize payments. Human approval remains the last step, highlighting current limits on agent autonomy in finance. Source: pymnts.com

NGMN: Telcos Need Guardrails Before Agentic AI Can Run Networks

According to NGMN, agentic AI could help telecom operators reach Level 4 network autonomy, but only after issues like identity management, governance, and interoperability are resolved. The group warns that without these foundations, deploying autonomous AI in network operations carries significant risk. Source: fierce-network.com

Amazon Bedrock Adds AgentCore Gateway to Control Agent Tool Access

AWS is introducing AgentCore Gateway for Amazon Bedrock, aimed at helping enterprises manage which tools their AI agents can access. The move responds to a common customer concern seen across coding and autonomous agent deployments regarding control and oversight. Source: aws.amazon.com

PEX Finance Chief Plans AI-Powered 'Shadow Ledger'

PEX CFO Luke Pritchett wants to deploy AI agents alongside the company's official books to build a parallel tracking system. The goal is to strip manual work out of his finance team's daily routine. Source: cfodive.com

Guide: Setting Up a Controlled Cursor MCP Workflow with Oracle AI Database

A new tutorial explains how to pair an Oracle SQLcl MCP server with Cursor CLI to manage database permissions and track tool usage. It also covers adding persistent memory for AI agents within this setup. Source: blogs.oracle.com

AI Agent Payments Are Coming—and They'll Reshape the Web

As AI agents bypass online ads entirely, the ad-supported internet business model faces disruption. New approaches like agentic commerce, pay-per-call systems, and the x402 protocol are emerging to enable payments between agents. Source: reports.tiger-research.com

Five Ways AI Agents Are Already Reshaping Industries

Agentic AI has moved beyond research labs into real enterprise deployments, according to a new roundup. The report says that by 2026, the industry conversation has shifted well past simple conversational chatbots toward practical, autonomous applications. Source: kdnuggets.com

Insurers Ramp Up Automation Spending as AI Agents Spread

Insurance companies are boosting investment in automation as they grapple with pressure on loss and expense ratios. The growing adoption of AI agents is a key driver behind this shift. Source: programbusiness.com

Ping Identity Extends Identity Security to AI Agents in Private Clouds

Ping Identity has launched self-managed identity controls designed to secure and oversee AI agents operating within private cloud and hybrid infrastructures. The feature targets organizations in regulated industries needing tighter governance over agent access. Source: scworld.com

Deloitte: Only 1 in 5 companies ready for autonomous AI agents

A new Deloitte report finds most organizations aren't prepared to shift toward autonomous AI agents. The main obstacles are poorly documented processes, fragmented data systems, and rigid organizational habits. Source: hrdive.com

Real-world AI agent failures rarely match vendor warnings

While vendors focus attention on model-level risks, one practitioner reports that in practice, serious failures in AI agent fleets more often stem from scheduling errors, faulty shell assumptions, expired credentials, and broken safety checks. Source: infoworld.com

20% of enterprises can't stop AI agents from overspending in real time

According to new VentureBeat Pulse Research, companies typically run three AI orchestration platforms simultaneously, yet 21% of them still lack the ability to halt an out-of-control agent's spending as it happens. Source: venturebeat.com