Salesforce has launched a rebuilt version of Slackbot, recasting Slack’s longtime workplace assistant from a lightweight notification feature into what the company describes as an AI agent for enterprise search, drafting, and task execution. According to VentureBeat, the new Slackbot is now generally available for Slack Business+ and Enterprise+ customers, with broader rollout continuing through the end of February and mobile completion expected by March 3.

The move matters beyond a single product refresh. Salesforce is trying to make Slack the place where employees ask questions, generate documents, assemble context from scattered systems, and eventually trigger actions across business software. That puts the company in more direct competition with Microsoft Copilot in Microsoft Teams and Google Gemini in Google Workspace, while also tying Slack more tightly to Salesforce’s broader pitch around AI agents and Agentforce.

For Salesforce, the launch also addresses a strategic pressure point. Investors have spent the last year asking whether generative AI will strengthen software incumbents or reduce them to infrastructure providers behind third-party assistants. Salesforce’s answer is that the chat layer itself can become the control plane for enterprise AI, provided it has enough context, permissions, and workflow reach.

A rebuilt Slackbot, not a small upgrade

VentureBeat reported that Salesforce executives described the product as a ground-up rebuild rather than an iteration on the old Slackbot. The earlier version handled reminders, notifications, and basic prompts inside Slack. The new system, according to Slack CTO and Salesforce co-founder Parker Harris, is built around a large language model, search, and connectors into enterprise data sources.

As presented to VentureBeat, Slackbot can search across Slack conversations, Salesforce records, Google Drive files, and calendar data. In a demonstration cited by the publication, Slack product staff showed the assistant analyzing customer feedback, correlating that material with an uploaded dashboard image, identifying possible accounts in Salesforce for a pilot, turning the output into a Slack Canvas document, and checking stakeholder availability for a follow-up meeting.

That workflow is important because it shows Salesforce is not positioning Slackbot as only a chat interface for Q&A. The company wants Slackbot to become a coordinating layer inside Slack Canvas and, over time, across external tools. Slack chief product officer Rob Seaman told VentureBeat that today’s document generation into Canvas is effectively an internal tool call and signals future third-party tool-calling plans.

If that roadmap materializes, Slackbot would move closer to a task-oriented assistant that works inside the user’s existing collaboration surface rather than a standalone chatbot tab. That is the same broad destination being chased across enterprise AI, but Salesforce is betting that Slack’s conversational history and channel structure give it an advantage in context.

Why Salesforce chose Claude first

According to VentureBeat, the new Slackbot currently runs on Claude from Anthropic. Harris said the initial model choice was shaped in part by compliance requirements, specifically that Slack’s commercial service operates under FedRAMP Moderate certification for U.S. federal customers and Anthropic was the compliant option available when the system was built.

That does not appear to be a long-term single-model strategy. Harris told VentureBeat that Salesforce expects to support additional providers this year and specifically mentioned Gemini as a strong option on both performance and cost. He also said OpenAI remains a possibility.

That is notable for two reasons. First, it suggests Salesforce is treating foundation models as swappable components under a product experience it controls. Second, it reflects a broader market shift in which application vendors increasingly want freedom to route tasks to different models based on latency, pricing, compliance, and quality rather than tie their products to one provider.

Salesforce also used the launch to make a familiar enterprise assurance point: Harris said the company does not train models on customer data. That statement addresses a common buyer concern, but it should be read as a vendor claim about product policy rather than independent verification of architecture or data handling practices.

The competitive fight is really about workflow ownership

The most obvious comparison is Microsoft Copilot, especially because Teams, Microsoft 365, Outlook, and Office documents already sit inside many enterprise workflows. Google Gemini has a similar built-in advantage across Workspace. Slack’s response is to argue that proximity and context inside day-to-day conversations can be just as powerful as document-suite integration.

Salesforce executives told VentureBeat that Slackbot’s edge is that it is already embedded where teams communicate and make decisions. The company also argues that Slackbot is grounded in the user’s existing channels, files, and permissions without requiring extensive setup by the end user.

That argument is plausible, but it also has limits. In organizations where Slack is the operational center, Slackbot may indeed be the fastest route to an assistive workflow. In organizations that live primarily in Microsoft 365 or Google Workspace, the opposite may be true. Much of the enterprise AI competition is becoming a battle over which surface is “close enough” to the work to become the default request layer.

Salesforce is also trying to widen the contest beyond assistants. Harris described Slackbot to VentureBeat as an eventual “super agent,” a central interface that could coordinate with other agents. That connects Slackbot to Agentforce, Salesforce’s larger vision for enterprise AI agents, and to external agent ecosystems emerging in Slack.

VentureBeat noted that Claude Code for Slack from Anthropic is already in preview, and that OpenAI, Google, and Vercel have also built agents for the platform. If Slack becomes a container for many specialized agents, Salesforce’s opportunity is not only selling one assistant but owning the environment where human workers and software agents interact.

Evidence, adoption claims, and open questions

Because this story is based on VentureBeat’s reporting from Salesforce interviews and product demonstrations, several of the strongest performance and adoption signals are vendor-reported and should be treated cautiously.

Salesforce told VentureBeat that it tested the new Slackbot internally with 80,000 employees. The company said two-thirds of employees had tried it, 80% of those users continued to use it regularly, and internal satisfaction reached 96%. Executives also said employees reported saving between two and 20 hours per week. Those figures may indicate strong internal interest, but they are self-reported or company-reported metrics, not independently audited usage data.

The same caution applies to pilot customer examples. VentureBeat reported comments from Beast Industries, Slalom, reMarkable, Xero, Mercari, and Engine. Those references are useful as early signals of market testing, but they do not establish broad enterprise traction. Reported time-savings figures such as 90 minutes per day for one Beast Industries employee and 30 minutes daily for an executive at Engine are anecdotal and role-specific.

There are also still product limitations. Seaman said calendar reading and availability checking are live, while meeting booking is expected a few weeks later. Image generation is not available. And Salesforce did not provide specifics to VentureBeat on support for competing CRM platforms such as HubSpot or Microsoft Dynamics. That omission matters because buyers will want to know whether Slackbot can act as a neutral enterprise layer or whether it works best when Salesforce is already the system of record.

Cost is another unresolved point. VentureBeat reported that Slackbot itself carries no extra charge for Business+ and Enterprise+ customers. But the publication also linked the launch to wider concerns around Salesforce data access pricing, including criticism from Fivetran CEO George Fraser that higher API-related costs could make it harder for enterprises to move Salesforce data into external systems like Snowflake or interact with it through tools outside Salesforce’s preferred stack. That issue is adjacent to Slackbot, but relevant: a product that promises broad enterprise context is more valuable when data movement is easy and less compelling when access becomes more constrained or expensive.

What this means for builders and enterprise buyers

For product teams and AI builders, the launch reinforces a clear design pattern: enterprise assistants are moving from generic chatbots toward context-aware agents embedded in collaboration tools. Slackbot, Microsoft Copilot, and Google Gemini are all trying to reduce the need for users to switch applications or manually gather source material.

For enterprise buyers, the key questions are operational rather than visionary. How well does Slackbot respect permissions? How reliably does it cite or ground answers in Slack and Salesforce records? How much setup is really needed for connectors and governance? And can it complete actions safely, not just produce polished summaries?

The Slack Canvas angle is especially relevant. Shared output inside a collaborative document can be easier to review, edit, and approve than one-off chatbot responses. If Slackbot becomes good at turning conversations into persistent artifacts and follow-up tasks, it could be more useful than assistants that stop at summarization.

For developers building on Slack, Salesforce’s comments about third-party tool calls and possible Model Context Protocol support are worth watching. A stronger agent interface inside Slack could create a new distribution path for niche enterprise actions, from code workflows to approvals to support operations. But that opportunity depends on how open Salesforce keeps the platform and how much it privileges its own products, including Agentforce.

What to watch next

The next useful signals will be concrete, not rhetorical. First, watch whether Salesforce adds the promised support for Gemini or other model providers, because that will show how serious the company is about a multi-model architecture.

Second, look for real-world evidence of tool calling beyond Slack Canvas, especially integrations that let Slackbot take actions in third-party systems with auditability and permission controls.

Third, track whether Salesforce clarifies support for non-Salesforce systems such as HubSpot, Microsoft Dynamics, Snowflake, or external data platforms. Buyers will want to know whether Slackbot is an enterprise assistant or primarily a Salesforce assistant living in Slack.

Finally, adoption outside Salesforce’s own workforce and curated pilot accounts will matter more than internal success stories. Usage retention, expansion within large customers, and evidence that security teams are comfortable with deployment at scale will be better indicators than executive enthusiasm.

Creati.ai perspective

Salesforce’s Slackbot launch is significant because it turns Slack from a communication product with AI features into a stronger contender for the enterprise AI interface layer. The company’s best argument is not that its model is uniquely powerful, but that work context already lives in Slack conversations, files, channels, and linked systems. If Slackbot can reliably turn that context into useful actions, Salesforce has a path to defend Slack against Microsoft Teams and make Agentforce feel practical rather than abstract.

But this is still an early product category wearing mature-software branding. The vendor-reported internal adoption numbers are encouraging, not definitive. The bigger test is whether enterprise customers see Slackbot as a trusted automation surface rather than just another assistant pane. In workplace AI, the winner is unlikely to be the system that writes the cleverest answer. It will be the one that fits permissions, workflow, compliance, and follow-through well enough to become part of daily operations. Salesforce has made a credible move with Slackbot. Now it has to prove it outside its own walls.