返回新闻
产品展示AI Understanding 简报

Google 扩展了 Gemini 的“Call for Me”功能以实现自动化业务任务

Google 正在推出一项功能,允许 Gemini 移动应用程序代表用户向企业拨打电话,目前仅限于美国号码。

4 min readRead the linked source
Source-page capture accompanying Google expands Gemini 'Call for Me' feature for automated business tasks
来源参考来源记录
出版商
support.google.com
来源链接
support.google.comhttps://support.google.com/gemini/answer/18336420?visit_id=639259251968538096-1942209662&p=lm_gemini_calls&rd=1
来源类型
链接来源——主要来源状态尚未确定。
背景60 秒内了解这一点

从这里开始

关键术语

特征
模型用来进行预测的输入变量。
检索
从知识源中查找相关文档或记录以进行查询。
测试一下自己AI 代理测验

发生了什么

Google has introduced a within the Gemini mobile app that enables the AI to place phone calls to businesses to perform tasks on a user's behalf. The functionality is currently being rolled out gradually to eligible users and is restricted to calls made to US phone numbers.

Google's 'Call for Me' allows users to request that the Gemini mobile app initiate phone calls to businesses to complete tasks. The company describes this as a tool for handling everyday errands, though it has not provided an exhaustive list of supported actions.

The rollout is described as gradual, meaning the may not be immediately accessible to all users. Google has explicitly stated that the service is currently limited to calling US phone numbers.

To address potential concerns regarding automated communication, Google has implemented an opt-out system that allows individuals to prevent their phone numbers from receiving calls initiated by other Gemini users. Users also have the ability to opt back in through the same settings interface.

The service includes daily usage limits on the number of calls an individual can initiate via Gemini. If a user hits this threshold, they must wait until the following day to resume using the .

来源详情: support.google.com ↗

为什么这很重要

This represents a shift toward agentic AI, where models move beyond information to executing real-world actions. By automating interactions with business phone systems, Google is testing the practical utility and social implications of AI-driven communication. The inclusion of an opt-out mechanism for receiving such calls highlights the emerging need for protocols regarding AI-to-human and AI-to-business interactions, as well as the potential for spam or misuse in automated telephony.

The integration of AI into telephony marks a significant step in the development of agentic AI, where the model acts as an intermediary in the physical world. This capability requires the AI to navigate interactive voice response (IVR) systems and communicate effectively with human representatives.

The introduction of an opt-out mechanism acknowledges the potential for friction in automated calling. As AI agents become more capable of performing tasks that were previously human-exclusive, the industry must establish norms for how these agents identify themselves and how recipients can manage or block such interactions.

Practical implications include potential time savings for users performing routine tasks, but also the risk of increased automated traffic to business support lines, which may necessitate new filtering or verification technologies for businesses.

Interactive Mechanism

互动机制:它实际上是如何运作的

以交互方式探索这一发展背后的基础技术。

Agent Lifecycle Stage:
1
User Intent & Planning: "Audit customer refund request #4092 and settle payment."
2
Tool Calling: Emits structured JSON call crm_get_transaction(id='4092').
3
Guardrail & Verification:🛡️ Paused: High-value action requires human operator sign-off.
4
Final Settlement: Refund recorded, email receipt dispatched, and audit log stored.
Core takeaway: An AI agent is not just a language model—it is a closed loop of planning, tool invocation, and environment feedback. Production systems require self-healing retries and strict human approval guardrails.
交互式概念检查+10 Points
AI Agents Quiz

An agent must create a draft calendar event for Tuesday at 2 p.m. Which evidence would establish the requested result?

接下来看什么

Users should monitor the expansion of this beyond US phone numbers and observe how businesses respond to AI-initiated calls. Key unknowns include the specific types of tasks Gemini is permitted to perform, the nature of the daily call limits, and how Google plans to verify the identity of the AI during these interactions to prevent fraud or confusion.

It remains unknown which specific business tasks are supported and which are explicitly excluded by Google's safety or operational guidelines.

The exact daily call limit per user is not specified, nor is the criteria for 'eligible users' who receive access during the rollout phase.

Future updates may clarify how the AI handles complex scenarios, such as verifying the user's identity with the business or managing sensitive information during a call.

相关指南和测验

人工智能代理AI 伦理ChatGPT 与大语言模型测试你所知道的——尝试免费的人工智能测验在我们的词汇表中查找人工智能术语关注 AI 模型发布跟踪器
觉得这有用吗?