Slack Launches Surfaces: Build Enterprise Tools in Chat Using Natural Language

Slack's new Surfaces feature lets users build interactive enterprise tools in chat using natural language.
Slack has launched Slackforce Surfaces, enabling users to build interactive reports, dashboards, polls, and more directly within the chat window using natural language — no external apps or coding needed. The feature uses AI to automatically aggregate data from conversation history and third-party apps like Salesforce and Google Drive, helping break down enterprise data silos. Key challenges around data accuracy, security permissions, and output controllability remain to be addressed.
Slack Steps Into the AI-Native Collaboration Era
Slack recently announced a new feature called Slackforce Surfaces, marking the platform's official entry into the "AI-native" era. Users will be able to build interactive reports, polls, dashboards, presentations, mini-websites, and more — all directly within the chat window, without switching to any external app.
The core idea behind this feature can be summed up as "vibe coding" — users simply describe what they need in natural language to Slackbot, and the AI automatically handles information gathering and tool creation. This aligns with the vibe coding concept that has recently gained traction in developer communities: building functionality through conversational interaction rather than traditional coding. What sets Slack apart is bringing this paradigm directly into the enterprise collaboration context.
How Surfaces Works
From Conversation to Finished Product
Slackforce Surfaces works in a remarkably intuitive way. After a user describes their needs to Slackbot, the system uses AI to automatically pull information from relevant conversation history and connected third-party apps. Supported integrations include major enterprise services like Google Drive and Salesforce.
Here's a concrete example: if a team lead wants to generate a quarterly performance dashboard, they simply describe that intent in natural language within Slack. The AI will automatically search through relevant team channel discussions, pull sales data from Salesforce, retrieve related documents from Google Drive, and compile everything into an interactive report interface — no manual data exports, no app switching, no coding required.
Breaking Down Enterprise Data Silos
The most valuable aspect of this feature is how directly it addresses the problem of enterprise data silos. In traditional workflows, a team's critical information is often scattered across chat logs, CRM systems, cloud storage, and various SaaS tools. Surfaces uses AI to automatically connect these disparate sources, transforming them into usable outputs right within the collaboration context.
It's worth noting that Slack, as a Salesforce subsidiary, has a natural advantage when it comes to deep integration with the Salesforce ecosystem. The "Slackforce" naming itself reflects the increasingly tight product convergence between Slack and Salesforce.
How Surfaces Changes Enterprise Collaboration
Dramatically Lowering the Bar for Tool Creation
Surfaces' most immediate value is bringing the barrier to creating internal enterprise tools down to an absolute minimum. In the past, building an interactive dashboard or mini-website often required IT department involvement or specialized BI tools — far out of reach for most business users. Now, any team member can generate these tools simply by describing their needs in natural language. In essence, this takes the "citizen developer" concept to a whole new level.
The diverse range of outputs — polls, presentations, mini-websites, and more — means the use cases are broad. From everyday team decision-making votes to client-facing project showcases to internal knowledge-sharing microsites, it can all be done in one place, right within the chat stream.
AI-Native Collaboration as the Industry's Big Bet
Slack's move is far from an isolated one. Notion, Microsoft Teams, Google Workspace, and other collaboration platforms are all actively embedding generative AI into their core workflows. Each takes a different angle: Microsoft leads with Copilot's general-purpose assistant capabilities, Notion focuses on AI generation within documents, while Slack carves out a unique position around "instantly building tools within conversational context."
Chat is already the most information-dense and real-time environment in team collaboration. By planting tool-building capabilities there, Slack is acting on a key insight: the moment information is generated is the same moment the need for a tool emerges. This design philosophy could well become an important direction for the future evolution of collaboration software.
Hurdles Still to Clear
Despite the promising picture Surfaces paints of automated collaboration, a few important questions deserve attention:
Data accuracy is the primary challenge. When AI automatically aggregates information from multiple sources, how can it ensure that generated reports and data are error-free? In enterprise decision-making contexts, a single wrong number can have serious consequences.
Data security and permission management are equally critical. When AI can automatically access data sources like Salesforce and Google Drive — both of which contain sensitive information — Slack must ensure that only authorized users can generate tools containing specific data.
Controllability of generated outputs is also key. While vibe coding lowers the barrier to use, it can also introduce unpredictability in what gets produced. Whether users can fine-tune and correct generated results will directly affect how usable this feature is in real-world work.
Conclusion
Slackforce Surfaces represents an important milestone in the evolution of enterprise collaboration tools toward an AI-native model. By bringing vibe coding into the chat context, Slack is trying to make building tools as simple as sending a message. If this feature delivers on its promise in practice, it has the potential to redefine how teams access information and produce outputs during collaboration. Whether it ultimately holds up will depend on how it performs on the critical dimensions of data accuracy, security, and output controllability once it's officially released.
Related articles

LangChain + MCP: From Core Concepts to Agent Tool Calling in Practice
Learn how LangChain and MCP work together — covering LLM tool calling, Agent architecture, and conversation history management to build real-world AI applications.

Probabilistic Machine Learning: Why It's the Cornerstone to Unlocking the ML Black Box
Without probability theory, ML is always a black box. This article explores why probabilistic foundations are essential for understanding machine learning algorithms, Bayes' theorem, MLE, and more.

Optimization Pitfalls in Self-Evolving LLM Agents: Value Concentration and Budget-Splitting Problems
HARNESSEVO research reveals 3 key LLM agent harness optimization findings: value concentrates in reflection/control slots, uniform budget splitting is harmful, and credit assignment must precede structured evolution.