The AI browser wars
There’s an interesting thing happening in the world of web browsers. While Google Chrome still dominates the traditional browsing experience, there is a whole crop of new browsers coming out from AI companies, with the goal of integrating their AI experience into the browser.
The idea is great, in theory, but comes at some costs with potential privacy and security concerns everywhere.
The first big one to come out was the “Comet” browser from Perplexity, and more recently we saw “Atlas” from ChatGPT (which I’m currently using a bit). There is also the Claude extension for Google Chrome if you use that app, and of course Chrome itself is slowly getting more integration from Gemini. Lastly, if you use the Brave browser (which is my primary one), they have an AI assistant named Leo in there.
So what do these browsers do?
Here is a pretty deep explanation of what ChatGPT’s Atlas does, as most of these browsers all work roughly the same:
To take it further, Casey Newton’s Platformer newsletter unpacks some of the good and the bad of it. As he shares, these browsers feel very much like Chrome (and they’re built on the same technology), they’re a push to simply get more use for their AI models, and he shares some of the security concerns.
The bigger question is the value (or lack of value) that this might bring. Is this a more useful browser than a standard one? As Newton shares, the potential for these kinds of products is huge (particularly the “Agent” modes, where it does work on your behalf), but we’re still a long way from that being as smooth as one might hope.
I encourage you to check out the video above, if for no other reason than to understand where things are going.
What is your main browser these days?
My main browser is Safari