What changes when OpenAI makes an AI choose from fixed answers?
OpenAI is previewing a service that chooses from developer-defined answers. What that changes for applications, and what its speed chart leaves unanswered.
AI Agents
How does Mistral want Vibe to remember work without overloading chat?
Mistral is turning Vibe into a unified work surface with editable knowledge pages and spreadsheet workflows. What does that change?
Can Trane's AgentCore assistant really make building diagnostics 60x faster?
AWS says Trane cut a building diagnostic workflow from 20 minutes to 20 seconds. What does the AgentCore case study actually show?
How does AWS want teams to test skill-equipped AI agents?
AWS is adding skill-focused evaluators for agents. What do they measure, and what remains unproven?
What is Google's Business Agent for YouTube Ads supposed to change?
Google says Business Agent for YouTube Ads will let viewers ask product questions beside video ads. What is actually new in the September Demand Gen Drop?
What does Gemini Live Avatar change for enterprise voice agents?
Gemini 3.8 Live with Live Avatar is now generally available in Gemini Enterprise. What changes for teams building voice and video agents?
Xiaomi opens MiMo-V2.6 for agentic RL tests
Xiaomi released and open-sourced MiMo-V2.6, pairing open weights with reinforcement-learning resources for agent, coding, cyber and multimodal tasks.