aifollow.news Search
Back IT之家 科技新闻
IT之家 科技新闻· · Original publication time Selected

Ant Group releases Lingguang 2.0, with tasks that continue after users leave a chatMachine translation

Automatically verified and published · Generated and evidence-checked automatically; not reviewed by a human.

AI-assisted summary

Ant Group has released Lingguang 2.0. The company says that, with user authorization, the assistant can identify needs using connected information and services and keep tasks running after a chat ends. For example, after connecting to Alipay, it can flag potentially forgotten auto-renewing subscriptions. Users can view, change or revoke permissions at any time.

Article · AI translation

IT Home reported on October 27 that Ant Group released Lingguang 2.0 today, advancing AI assistants from passively responding to instructions and completing one-off tasks toward proactively identifying needs and continuously managing matters.

According to Ant Group’s official introduction, a major change in Lingguang 2.0 compared with traditional conversational AI and task-oriented agents is that AI becomes involved at an earlier stage of a task.

Previously, users had to identify a problem and state their needs before AI could act. Now, with user authorization, Lingguang can draw on connected information and services to proactively identify potential needs, initiate analysis, monitoring and follow-up, and promptly alert users when their attention is needed. AI is no longer merely a tool waiting for instructions; it is beginning to become an assistant capable of managing matters continuously.

Behind this change is a comprehensive upgrade to Lingguang 2.0’s cross-session memory, long-running task scheduling and ability to connect with tools. Using a dedicated cloud host, Lingguang can move beyond the limits of a single conversation, continue running tasks after a user leaves the conversation, and follow up dynamically as new information emerges and circumstances change, shifting AI from one-off task execution to long-term, continuous management of matters.

Lingguang’s ability to manage matters is also entering specific areas of daily life.

In consumer spending, after a user authorizes access to Alipay, Lingguang can review transaction records, identify auto-renewing subscriptions that may have been forgotten, analyze spending habits and offer money-saving suggestions, among other things;

In travel, AI service does not end once flights and hotels have been booked. Lingguang can continue tracking subsequent price changes, spot opportunities to pay less and notify users promptly;

In family life, once authorized, Lingguang can monitor teachers’ notices in DingTalk class groups, organize important information, extract to-do items and handle follow-up work such as document formatting.

Lingguang can also, within the scope of its authorization, proactively identify key dates and arrange follow-up for matters requiring long-term memory and ongoing attention, from pet care and childhood vaccinations to family anniversaries and annual vehicle inspections.

According to the introduction, Lingguang 2.0 connects data and services within the Alipay ecosystem and also supports the use of tools such as MCP and CLI in open environments, expanding its ability to coordinate across platforms. Lingguang currently has built-in capabilities for Alipay bills, memberships and travel, as well as DingTalk, Feishu, email and Mi Home. The scope of what it can manage is expanding from personal information and online matters to cross-platform workflows and smart-home device control.

As agents move toward “managing matters,” the limits of their permissions, data security and the trustworthiness of their actions are also becoming new challenges. IT Home learned that Lingguang 2.0 has integrated Ant APASS. As a unified account and security management infrastructure for the AI era spanning agents, devices, applications and service scenarios, APASS incorporates identity authentication, intent security and privacy protection into the agent execution process. Through mechanisms such as separate custody of authorization keys and agents, it enables agents to carry out tasks without directly accessing authorization keys. Users can view, adjust or revoke authorizations at any time, keeping agents’ actions within the authorized scope.

Found an error? Send a correction