Computer in Slack: From Shared Context to Finished Work

AI delivers the most value when it works inside the tools where teams already communicate and collaborate.

Today we’re launching Computer in Slack, which lets users kick off tasks with the powerful AI agent orchestrator directly in the flow of work. 

In Slack, teams can use Computer to search the web, work across connected tools, generate content, analyze business data, and build outputs like apps and websites. They can also ask Computer to run complex workflows directly from Slack.

Computer in Slack keeps the query, the context, and the output together, so teams can collaborate with Computer over time all in the same thread.

Started as an internal tool

Like many breakthrough technologies, Perplexity Computer began as an internal tool at Perplexity. Computer was born in the place where we do most of our work: Slack. 

At first, we called it ASI, a play on artificial superintelligence and “a Slack integration.” In the channel employees could simply tag ASI, and ask it to perform any task you can think of. That included pulling data from Snowflake, researching the web, preparing board memos, writing code, and more. 

Soon, we realized with deep research and web access, it was much more than ASI. So we renamed it Computer.

In its first four weeks of existence as a Slack integration, Computer performed $1.6M of work for 300 Perplexity employees. But the numbers only told part of the story. 

Because we were using it in a public Slack channel, everyone could see better ways to use Computer, and tag Computer into more shared workflows. Collaboration improved Computer. 

We began to work on a public version of Computer, but we knew Slack would always be a powerful place to use it.  

Collaboration requires context

When working in Slack, teams can keep the query, the context, and the output together. Teammates can add detail in the thread, project leads can hand off work, and everyone can review the result in the same shared workflow. 

The work stays in one place instead of moving across separate tools.

What teams use it for

Users can work with Computer directly in Slack in a lot of different ways. For example, users can ask it to:

• “Review this partnership proposal, fact-check the statistics, flag risky terms, and return a marked-up Word document with suggested edits.”

• “Draft this quarter’s OKR kickoff deck using our strategy docs, prior results, and current backlog.”

• “Find the sales call appointments that showed up in GoHighLevel, review the recordings, and send a summary report to the sales manager in Slack.”

How to use Computer in Slack

When a task belongs in the open, tag @Computer in the message box in a shared channel or group conversation and describe what’s needed in plain language. That works best when other people need visibility or want to add context. 

When the work is more individual, send Computer a direct message. That is a better fit for first drafts, quick research, or tasks that need shaping before bringing them back to the team.

Computer may ask follow-up questions before it starts. That usually means it is clarifying the goal or the output. Once it has what it needs, it can keep working in the background while you move on to something else.

Move between Slack and Perplexity 

Everything queried in Slack can also be seen in the Perplexity web platform. The experience is mirrored, so a task can start in a shared channel and continue later on the web without losing the thread.

The new MCP connector gives Computer deeper access to Slack context and actions. It can search workspace knowledge, read threads, send messages, and stay better grounded in the work already happening there.

Now available 

Perplexity Computer is easy to install directly as a Slack integration. Try Computer in Slack today and start where the work already is. 

Computer is also now available in the Slack Marketplace.