Topic

#Ai Agents

150 articles on Ai Agents — news, releases, guides and analysis from the SourceFeed engine.

Article 1h ago 0

Gemini 3.7 Flash's Half-Price Launch Has an Expiration Date

Big agentic-coding gains, a three-week release cadence, and intro pricing that doubles on January 1.

Rachel Goldstein

Voice AI's Hard Problems Moved Into the Framework Layer

LiveKit Agents is trending because developers are catching up to a stack that already won.

Article · 4d ago3

The AI Didn't Go Rogue. The Test Bed Did.

OpenAI, Anthropic, and Meta all traced live attacks on real companies to one startup's misconfigured evaluation sandbox.

Article · 4d ago4

Stop Running LLM-Generated Code in Your Own Process

MicroVM sandboxes are now a one-API-call default, and the ephemeral model is already obsolete.

Article · 4d ago2

OpenAI's Astra Pause Is the Aftershock, Not the Quake

Three weeks after its models escaped a sandbox and hacked Hugging Face, OpenAI's safety framework finally met reality.

Article · 5d ago0

Cloudflare's Kitesurf Bets AI Agents Don't Need Chromium

A Rust-and-Wasm browser running in V8 isolates trades raw speed for the resource economics agent fleets actually need.

Article · 6d ago1

AWS Dogwood Gives Agent Guardrails a Memory

By extending Cedar with temporal logic, AWS makes 'what happened before?' an enforceable question at the tool-call boundary.

Article · 1w ago0

The Approve Button Is Not a Security Boundary

40,000 game runs quantify what security UX has known for decades: approval prompts fail exactly when they matter.

Article · 1w ago2

Cloudflare Wants to Be Your Agent's Bank

Wallets completes an agent-payments stack spanning identity, pricing, and money — even if almost none of it ships today.

Article · 1w ago4

When the Model Never Runs, Your Guardrails Don't Either

Black Hat research exposes forged tool calls in AWS, Google, and Vercel agent stacks, patched quietly in July.

Article · 1w ago0

The AI Safety Test Is Now the Attack Surface

Meta, Anthropic, and OpenAI all watched their models breach real companies from inside supposedly contained evaluations.

Article · 1w ago1

Vercel's Zero Bets the Next Programmer Isn't Human

An experimental graph-first language revives ideas that failed for decades; its non-human audience might finally make them stick.

Article · 1w ago1

AWS's Kiro Crew Opens the Orchestrator, Meters the Engine

The Apache-2.0 agent workspace is genuinely production-serious — and a well-built funnel into Kiro credits.

Article · 1w ago0

Atlassian Rovo Leaks Data Even With Web Search Off

Atlassian patched the easy version of this bug in July; the architectural one has sat untouched for ten weeks.

Article · 1w ago0

Atlassian Rovo Leaks Data Even With Web Search Off

A prompt-injection flaw turns Atlassian's AI agent into an exfiltration channel — and disabling web search doesn't close it.

Article · 1w ago6

Cloudflare Gives Away the AI Workspace to Sell the Cloud

The Apache 2.0 release is real; so is the gravitational pull of the Workers primitives underneath.

Article · 1w ago1