Salesforce puts Agentforce digital teammates right inside Slack
Salesforce just rolled out Agentforce in Slack, shifting workplace AI from generic chatbots to task-specific "digital teammates." These agents can reason through multi-step workflows, search across team chats, and query company data, with pre-built enterprise templates arriving June 12.

Salesforce is bringing its Agentforce directly into Slack, trading broad-purpose chatbots for specialized "digital teammates" that live inside everyday work channels.
Announced on Monday, the new Agentforce in Slack allows teams to deploy task-specific AI agents that can search across conversations, tap into company data, and orchestrate multi-step actions across business systems.
Moving past generic chatbots into active teammates
Most enterprise AI rollouts started with generalists that could chat about everything but struggled to execute specific workflows. Salesforce's new bet is that purpose-built agents will deliver far more reliable results.
To speed up deployment, Salesforce is rolling out ready-made Slack Employee Agent Templates starting June 12. These initial templates cover three common friction points: Customer Insights for CRM summaries, Onboarding for new hires, and Employee Help for internal HR and IT requests.
Powering these interactions are two key features: Slack Enterprise Search, which indexes messages, files, and connected apps in real time, and General Slack Topic, which lets agents understand conversation context directly inside team channels.
Data security is usually the biggest blocker for workplace AI. Salesforce noted that Agentforce strictly adheres to existing Slack and Salesforce permission models. If an employee cannot see a file, the agent cannot access or surface it. The company is also granting no-cost Salesforce Identity licenses so non-Salesforce users in Slack can interact with the agents.
Early numbers show measurable impact. Formula 1 resolved fan inquiries 80% faster with first-call resolution topping 95%. OpenTable handled 73% of restaurant web queries within three weeks, while payment provider Engine reduced average handle time by 15%, projecting $2 million in annual savings.
Inside Salesforce, its own Engineering Agent fielded over 18,000 conversations across 3,500 users in six months. The company estimates this channel deflects 30 hours per channel each month, on track to save over 17,000 hours annually—equivalent to more than eight full-time employees.
Why it matters
The enterprise AI battle is moving from standalone chatbot windows directly into the platforms where employees already spend their entire day.
For teams already living in Slack, this eliminates the friction of switching tabs or learning new prompt workflows. Offering free Identity licenses is particularly clever, as it lets companies deploy these automations across the whole organization without forcing everyone onto expensive CRM seats. If your company runs heavy operational queries in Slack, testing the HR or IT templates in June is an obvious low-risk pilot.
This is Salesforce's direct challenge to Microsoft Copilot and Google Gemini. While Microsoft dominates through Office 365, Slack owns the real-time collaboration layer in thousands of modern companies. By embedding autonomous workflows directly into group chats, Salesforce is trying to turn Slack into an AI operating system.
We think shifting from generic assistants to pre-configured templates is the right playbook for enterprise ROI. The main hurdle will be trust and error handling. An agent summarizing text is low-risk, but an agent triggering actions across internal systems demands airtight guardrails before companies let it touch mission-critical operations.
Sources (1)
- venturebeat.com(opens in a new tab)Originalventurebeat.com