Skip to content
Tech FrontlineBiotech & HealthPolicy & LawGrowth & LifeSpotlight
Set Interest Preferences中文
Tech Frontline

The Rise of Agentic AI: Perplexity and Read AI Launch Autonomous Task Orchestrators

Perplexity and Read AI have launched autonomous 'agentic' tools capable of orchestrating tasks independently. ServiceNow reports resolving 90% of IT requests via agents, signaling a shift from AI chat to AI action.

Jason
Jason
· 5 min read
3 sources citedUpdated Feb 27, 2026
A conceptual diagram showing a central 'Master AI' brain glowing in blue, sending out smaller nodes

⚡ TL;DR

Agentic AI tools from Perplexity and Read AI can now manage tasks and other agents autonomously, redefining office productivity.

The Rise of Agentic AI: Perplexity and Read AI Launch Autonomous Task Orchestrators

The Shift to Autonomous Execution

The AI industry is moving beyond the "Chatbot" era into the age of "Agentic AI." Today, Perplexity and Read AI unveiled powerful new systems designed to execute complex workflows autonomously, shifting the paradigm from information retrieval to active task management.

According to Ars Technica (2026), Perplexity has launched "Computer," a revolutionary agent designed to act as a manager for other AI agents. It can take a broad goal, break it into sub-tasks, and orchestrate other specialized agents to complete the work with minimal human intervention.

Digital Twins and the End of Busywork

Read AI has introduced "Ada," an email-based digital twin. As detailed by TechCrunch (2026), Ada is capable of managing an executive’s inbox, negotiating meeting times, and extracting answers from company databases to respond to inquiries independently.

The scale of this shift is further validated by ServiceNow. VentureBeat (2026) reports that ServiceNow is now resolving 90% of its own internal IT requests autonomously, achieving resolution speeds 99% faster than human-led processes.

Future Outlook: A New Operating System for Work

This transition is supported by research into Multi-Agent Systems, such as Toward Expert Investment Teams (ArXiv 2026), which shows that fine-grained task decomposition improves AI reliability. The rise of agentic tools suggests a future where work is managed at the objective level, with an orchestration layer handling the software for you.

FAQ

什麼是「代理型 AI」(Agentic AI)?

是指具備自主執行能力的 AI,不只是回答問題,還能自行使用工具、規劃步驟並完成複雜任務,甚至能指揮其他 AI。

Perplexity 的 「Computer」 功能有什麼用?

它像是一位 AI 經理,能接收一個大目標,然後將其拆解,並指派給最適合的微型 AI 代理去完成,最後彙整結果。

Read AI 的 「Ada」 真的能幫我回電子郵件嗎?

是的,它被設計為電郵型的「數位孿生」,能根據你的空檔安排會議,並依據公司知識庫自動回覆常見問題。

📖 Sources