I don't believe it does give them your credentials? It's more like running something like AutoHotkey and having it send screenshots to Perplexity and inject whatever input events they send back, but integrated directly into the browser instead of running as a separate program.
They are not interacting with Amazon at all. They are sending control commands to your computer, where you've opened a session logged into Amazon as you. Amazon is claiming they can't run commands on your computer.
The whole thing becomes entirely absurd if you consider that a cloud model isn't even fundamental to the setup; you could just as well slot in a local model or codex or anything else and the scenario would be exactly the same. Perplexity is just acting as a model provider here.
I know it feels like they're different, but I think that at its core, they're the same.
When you say I am "manually controlling the site", that actually means I run a program on my computer that makes requests to the site, chooses how to display the content provided by the site, and then gives me control mechanisms (e.g. clicking, scrolling) to interact with that content.
An "agentic workflow" does the same thing. It's a computer program that I run that makes requests to the site, chooses how to display the content provided, and then gives me control mechanisms (e.g. typing text) to interact with the content.
I'm not trying to be pedantic here. I really think that the two aren't that different.
Comet IS a normal chromium browser. The assistant thing is just a chat on the right sidebar and you can tell it stuff like “click around on this website to do X”, which it can then do for a limited amount of turns
> What about giving it to your secretary with a list of what to shop?
"When a Comet user directs the Assistant to locate an item on Amazon.com, the Assistant takes screenshotsof the browser view, sends those screenshotsfrom the user’s computer to Perplexity’s servers, and receives instructions from Perplexity’s servers on how to navigateAmazon.com. In other words, the Assistant cannot operate wholly independently; it relies on direction from the user and instructions from Perplexity’s servers."
Giving perplexity your credentials to run agentic workflows is not the same as manually controlling the site.
It’s literally not the same. I’m not really sure what else to tell you.
I don't believe it does give them your credentials? It's more like running something like AutoHotkey and having it send screenshots to Perplexity and inject whatever input events they send back, but integrated directly into the browser instead of running as a separate program.
They are not interacting with Amazon at all. They are sending control commands to your computer, where you've opened a session logged into Amazon as you. Amazon is claiming they can't run commands on your computer.
The whole thing becomes entirely absurd if you consider that a cloud model isn't even fundamental to the setup; you could just as well slot in a local model or codex or anything else and the scenario would be exactly the same. Perplexity is just acting as a model provider here.
I know it feels like they're different, but I think that at its core, they're the same.
When you say I am "manually controlling the site", that actually means I run a program on my computer that makes requests to the site, chooses how to display the content provided by the site, and then gives me control mechanisms (e.g. clicking, scrolling) to interact with that content.
An "agentic workflow" does the same thing. It's a computer program that I run that makes requests to the site, chooses how to display the content provided, and then gives me control mechanisms (e.g. typing text) to interact with the content.
I'm not trying to be pedantic here. I really think that the two aren't that different.
Comet IS a normal chromium browser. The assistant thing is just a chat on the right sidebar and you can tell it stuff like “click around on this website to do X”, which it can then do for a limited amount of turns
It’s not using Codex to steer a headless browser
What about giving it to your secretary with a list of what to shop?
> What about giving it to your secretary with a list of what to shop?
"When a Comet user directs the Assistant to locate an item on Amazon.com, the Assistant takes screenshotsof the browser view, sends those screenshotsfrom the user’s computer to Perplexity’s servers, and receives instructions from Perplexity’s servers on how to navigateAmazon.com. In other words, the Assistant cannot operate wholly independently; it relies on direction from the user and instructions from Perplexity’s servers."
2 replies →
[dead]