Salesforce Launches Agentforce 360 Amid Enterprise AI Race
▼ Summary
– Salesforce launched Agentforce 360, an updated AI agent platform featuring new tools for instruction, deployment, and Slack integration.
– The platform includes Agent Script, a prompting tool in beta from November that enables AI agents to handle flexible “if/then” scenarios.
– Agentforce Builder allows users to build, test, and deploy AI agents from a single location, also entering beta in November.
– Salesforce is integrating Agentforce with Slack, with core apps appearing in Slack starting this month and expanding through early 2026.
– Despite enterprise AI challenges, Salesforce claims 12,000 Agentforce customers, with early pilot users including Lennar, Adecco, and Pearson.
Salesforce has introduced Agentforce 360, a significant upgrade to its AI agent platform, as competition intensifies in the enterprise artificial intelligence sector. The announcement precedes the company’s Dreamforce customer conference starting October 14. This enhanced platform delivers several new capabilities, including text-based methods for directing AI agents, a unified environment for constructing and launching agents, and improved infrastructure for the Slack messaging application.
A key innovation within Agentforce 360 is the Agent Script prompting tool, scheduled for a beta release in November. This feature enables users to program their AI agents with greater flexibility, allowing them to handle conditional “if/then” scenarios more effectively. This programming makes AI behavior more predictable, especially when managing variable customer inquiries.
The platform also incorporates “reasoning” models, designed to process information before generating a response rather than relying purely on pattern recognition. These sophisticated agents are powered by underlying technology from Anthropic, OpenAI, and Google Gemini.
Another major component is the Agentforce Builder, a new tool that provides a single location for building, testing, and deploying AI agents. Also entering beta in November, this builder includes Agentforce Vibes, an enterprise-grade application for coding agent personalities that Salesforce revealed earlier this month.
A deeper integration between the Agentforce platform and Slack represents another strategic move. Core applications such as Agentforce Sales, IT, and HR will begin appearing directly within the Slack interface this month, with a full rollout expected by early 2026.
Simultaneously, Slack is testing a new iteration of its Slackbot. This version is evolving into a personalized AI assistant that learns user preferences to provide tailored insights and proactive suggestions.
Looking further ahead, Salesforce envisions Slack functioning as a central enterprise search hub. The company plans to introduce connectors for platforms including Gmail, Outlook, and Dropbox in early 2026.
This product launch occurs during a dynamic period for enterprise AI. Many technology firms are aggressively marketing AI solutions to corporate clients, yet businesses frequently report challenges in achieving a clear return on investment from these tools.
The competitive landscape is heating up. Recently, Google launched Gemini Enterprise, a collection of tools for developing enterprise-grade AI agents, with early adopters like Figma, Klarna, and Virgin Voyages. Similarly, Anthropic is gaining momentum with its Claude Enterprise product, securing a major deal with the global consulting firm Deloitte to deploy its chatbot to 500,000 employees, its largest enterprise agreement to date. A strategic partnership with IBM followed shortly after.
Salesforce emphasizes its strong market position, claiming 12,000 customers for its Agentforce platform, a figure it states surpasses all competitors. Early adopters testing the new Agentforce 360 features include Lennar, Adecco, and Pearson.
This commercial activity persists even as research, such as a recent MIT study, indicates that a staggering 95% of enterprise AI pilot projects fail to progress to full-scale production, highlighting the ongoing corporate struggle to validate expenditures on artificial intelligence tools.
(Source: TechCrunch)
