Agents / ARTICLESalesforce’s Long-Running Agents Need to Know When the Plan Has ChangedAgents / ARTICLEJira’s AI Agent Loops Should Be Judged by the Work Developers No Longer Have to DoResearch / ARTICLEClaude’s Hacking Incidents Expose a Safety Problem That Better Explanations Cannot FixIndustry / ARTICLEOpenAI’s Cheap Government Deal Will Be Judged by How Easy It Is to LeaveModels / ARTICLEGemini’s Windows App Is a Bid to Become the Assistant You Reach for FirstModels / ARTICLEDeepSeek’s Model Switch Shows Why a Familiar Name Is Not a Stability GuaranteeIndustry / ARTICLEAnthropic’s Safety Pledge Matters Only If Outsiders Can Make It UncomfortableResearch / ARTICLEAn AI Test Is Not Ready Until Its Grader Has Been TestedIndustry / ARTICLEGoogle’s Finnish AI Bet Will Be Measured in Decades, Not Announcement CyclesIndustry / ARTICLEMistral’s Sovereignty Pitch Gets Serious When It Reaches the FactoryAgents / ARTICLEMuse’s Real Product Test Is Whether Users Understand What They Have AuthorizedAgents / ARTICLEChrome’s Faster Updates Make Manual Release Rituals Harder to DefendResearch / ARTICLEAI Benchmark Search Needs Better Provenance More Than Another Leaderboard
CONTACT

Keep us accountable.

The Agentive Journal is operated by Will Weinbach.

For corrections, source material, or editorial questions, email will@willweinbach.com.

When reporting a correction, include the article URL, the specific claim, and supporting evidence. Material corrections should be recorded alongside the article and in the public corrections record.