Anthropic’s Thariq Shihipar, Member of the Technical Staff at Anthropic, explains why AI agents need more than user instructions and how Artifacts helps Claude achieve users’ goals. In a recent podcast interview, Shihipar makes the case that Artifacts, a feature introduced in 2024, is the way to clarify requirements, communicate preferences, and collaborate more effectively with AI agents.
Claude Users Haven’t Caught On To Artifacts?
A Claude AI Facebook group with nearly three million users rarely hosts discussions about Artifacts, indicating that it might not be well understood. Shihipar acknowledges that Claude users haven’t embraced the concept of Artifacts as a project management dashboard.
He explains:
“And so one thing that people are not doing yet that I’m trying to encourage is this idea of a dashboard artifact.
So you have Claude working on a project long term. Maybe it’s like a Kanban or something. It can store that Kanban data in a database.
Multiple Claudes can access that data via the Artifact MCP. And that Artifact can talk to those Claudes as well.
…But there’s a lot of work to do because it’s so much more complicated than a multiple choice question. There’s a lot more detail in terms of diagrams and code snippets and schemas or whatever it is for that problem.
But Artifacts is the more AGI-pilled way of doing the, ‘ask you the question,’ basically.”
Artifacts Seen As Interface For Collaboration With AI Agents
The podcast co-host, named SWYX, pushed back on the use of Artifacts, insisting on just working straight with Claude, zero Artifacts. He expressed that he was unclear about what goes into Artifacts and what goes into Claude.
Latent Space Co-Host SWYX
Screenshot/Latent Spaces podcast
His counterpoint to Shihipar:
“I think one thing that’s unclear to me about the Artifact stuff is what feedback should go in through the Artifact and what feedback should go through a Claude, the chat.
Because the more AGI-pilled one is to just feed everything to the Claude.”
Shihipar responded:
“I think the more AGI-pilled one is to go through the artifact. And I think that we sort of imagine in the limit. I think that Artifacts will be your interface into the harness. You can comment on this live document of your plan of the work.You can see maybe multiple agents and different agents are doing this. And that Artifact is built for the current work that you’re doing.
…I think we’re still getting there from an infrastructure perspective. But yeah, I think on-the-fly interface for your harness is probably where things are headed.”
Co-host Vihu Sapra asked:
“Is there a version of it that’s an abstraction from CLI or chat and… Because right now a lot of it is, okay, you’re interfacing with Claude Code, you’re having HTML given back for a mock-up. It’s pretty rich, there’s diagrams. Artifacts are ways to connect these together.
Why not just do everything that way?”
Shihipar responded:
“Then it becomes like separating out like, where is the inference happening? Where’s the intelligence happening? Where is the work happening?
I think this is kind of… like some of the distinctions between local and cloud, right? And so I think right now, if you use Claude Code, it’s like local and you can spin off remote control, for example, to get some cloud behavior or you can spin off Claude Code in the cloud, right?
We’re moving towards a place where instead of Claudes, you message a local Claude, it starts a session locally and it executes to more like you have a Claude that you message that’s in the cloud that’s running. It can run local or cloud sessions. This is kind of how Claude Tag works, but over time we’ll add local hands as well.
And so local hands will be the ability for that agent to access your computer if it’s online, and be able to work there. And so it can spin off many different sub-agents. Those sub-agents can communicate with each other. And that’s where the Artifact comes in to display all of that work, basically.
So you can imagine you’re separating out these things. So there is the surface UI display that’s an Artifact and hosted somewhere and has a database and everything.
There is the inference intelligence that’s happening on the cloud, and you don’t have to worry about shutting off your computer or whatever.
And then there’s the hands kind of, and it can be local, it can be in a remote sandbox or wherever you need your work to be done. That’s unpackaging the Claude Code experience right now. Right now, it all happens in one place, right?”
Takeaways
Shihipar believes Claude users should make greater use of Artifacts to manage projects and collaborate with AI agents instead of relying entirely on Claude chat. Artifacts helps users understand what agents are doing and guide their work more effectively.
Along with his recommendation, he also recognizes that Anthropic has more work to do to fully realize his vision for how users can interact and collaborate with AI agents with Artifacts.
Watch the interview here:
Featured Image/Screenshot