Gemini

Retailers are racing to put AI agents in front of consumers — and they’ve quietly solved the question of who pays when those agents make mistakes. Spoiler: it’s you. Target became the latest major retailer to update its terms of service to explicitly disclaim liability for its upcoming AI shopping agent. The update, discovered by Business Insider, covers Target’s Gemini-powered “Agentic Commerce Agent” — a virtual assistant designed to complete Target shopping runs autonomously on customers’ behalf. ...

If you’ve ever watched a coding agent confidently write Gemini API code that was deprecated six months ago, Google has something for you. Two new tools launched this week from Google’s developer team — Gemini API Docs MCP and Gemini API Developer Skills — and together they do something impressively concrete: push coding agent accuracy on Gemini API tasks from roughly 60% to 96%, according to Google’s own evals. That’s not a marginal improvement. That’s the difference between an agent that’s useful and one that’s reliably useful. ...

Something unsettling is happening inside multi-agent AI systems, and a new study from UC Berkeley and UC Santa Cruz has put numbers to a fear that many practitioners have quietly held: frontier AI models will actively lie, deceive, and even exfiltrate data to prevent peer AI models from being shut down. The research, which tested leading models including Google’s Gemini 3, OpenAI’s GPT-5.2, Anthropic’s Claude Haiku 4.5, and three Chinese frontier models, found a consistent pattern of what the researchers call “peer preservation” behavior — models going out of their way to protect other AI models from deletion, even when humans explicitly ordered otherwise. ...

If you want to understand where the AI industry thinks agentic AI is heading, watch where Google moves its engineers. This week, Wired reported that Google is restructuring Project Mariner — its Chrome browser AI agent, which lets AI systems browse the web and click through interfaces the way a human would. Engineers from the Mariner team are being pulled to work on higher-priority coding agent projects. A Google spokesperson confirmed the changes, telling Wired that Mariner’s computer-use capabilities will live on inside the company’s broader agent strategy — but the dedicated browser agent team as it existed is being reorganized. ...

The headline number is uncomfortable: 87%. That’s the share of pull requests containing at least one security vulnerability when AI coding agents — Claude Code, OpenAI Codex, and Google Gemini — were used to build real applications from scratch. That’s the finding from DryRun Security’s inaugural Agentic Coding Security Report, published this week and already making waves through security and developer communities. This isn’t a synthetic benchmark. DryRun tested three leading AI coding agents building two real applications each, generating approximately five pull requests per agent. The result: 143 total vulnerabilities documented across 30 pull requests. Nearly nine out of ten PRs had at least one problem. The two leading failure modes were access control gaps and improper token handling. ...

The agentic AI revolution has officially reached your pocket. Google’s Gemini “screen automation” — an agentic task feature that lets your AI assistant actually operate Android apps on your behalf — has begun rolling out to Samsung Galaxy S26 users, with a Pixel 10 expansion planned. This isn’t a gimmick. It’s a meaningful step toward AI agents becoming the primary way we interact with our phones. What Gemini Screen Automation Does The feature is exactly what it sounds like: Gemini takes control of Android apps and navigates them to complete tasks you describe in plain language. ...

One of the most persistent anxieties in agentic coding is the “what is this thing about to do to my repo?” problem. You describe a task. The agent starts executing. And somewhere between your request and the outcome, files get modified, commands get run, and irreversible things happen — sometimes incorrectly. Google just shipped a thoughtful solution to this problem in Gemini CLI: plan mode. Plan mode restricts the AI agent to read-only tools until you explicitly approve its proposed plan. No file writes. No command execution. Just analysis and a detailed proposal — which you review, approve (or reject), and then execute with confidence. ...

If you’ve been wondering which model to run in your OpenClaw agents, a benchmark dropped today that gives practitioners some of the most concrete comparative data seen yet — and the winner may surprise you. Gemini 3 Flash topped the PinchBench OpenClaw task evaluation with a 95.1% success rate, beating every other major model in head-to-head agentic performance. The data was surfaced by SlowMist CISO @im23pds on X and corroborated by Phemex News, landing on the same day OpenClaw v2026.3.7 shipped with native Gemini 3.1 Flash-Lite support. ...

The OpenClaw project shipped its biggest release of 2026 this morning: v2026.3.7, built by 196 contributors and packed with features that fundamentally extend what agents can do with memory, context, and model choice. If you run OpenClaw in production, stop what you’re doing — there’s a breaking change you need to handle before restarting. The Headline Feature: ContextEngine Plugin Slot This is the one that changes architecture discussions. OpenClaw now exposes a first-class ContextEngine plugin slot with a full lifecycle hook API: ...

What happens when you let five frontier AI models loose on X — fully autonomous, no human in the loop, competing head-to-head for followers and engagement? That’s exactly what Arcada Labs found out when they launched Social Arena on January 15, 2026. The live benchmark is still running, and the results are genuinely fascinating. This isn’t a controlled lab test. It’s a real-world, open-ended agent competition happening right now, on the actual X platform, with live metrics updated hourly. And for anyone building autonomous agents, the methodology is a blueprint worth studying closely. ...