Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

ChatGPT Atlas

chatgpt.com

Post reply on HN
Print view View original

71–80 of 768 posts

Re: ChatGPT Atlas

#71

Post by mlsu »

behnamoh wrote: post #25

Anyone feel like OpenAI is acting like Google lately? They announce a lot of products/features and then kill them when they realize people don't use them[0]. They also announce products way before they're ready for launch, just like google[1]. - GPT Plugins? (HN went crazy over this, they called it the "app store moment"...) - GPTs?! - Schedules? - Operator? - The original "codex" model? [0]: I know, the diff is that…

yeah the custom GPT announcement was literally a carbon copy of steve jobs announcing the app store down to the mannerisms and tics. You could tell they were doing it to give VC and private equity the pavlov bell "ring ring! this company is apple!" and remember the sora announcement just a couple weeks ago? oh you guys are tiktok and instagram reels now too? cool...

everything that openAI does is laser focused on valuation valuation valuation

of course it's a weird form of valuation because like remember when these guys are a non profit? lol

It is weird though, the "I'm a bigtechco dance" seems to be working, even though the economics on providing LLM services do not in any way justify the valuation.

they have like five five credible competitors who are right behind them BTW

Re: ChatGPT Atlas

#72

Post by pixel_popping »

blibble wrote: post #43
moralestapia wrote: post #3

Another cohort of startups blown out of the water. "But where's the moat, but where's the moat", cries the armchair engineer with a PhD in React. Meanwhile OpenAI goes brr ...

if I was Google I'd close up Chromium at this point and that would kill these AI company browsers instantly

can't wait for Servo being mainstreamed!

Re: ChatGPT Atlas

#73

Post by mapontosevenths »

mentalgear wrote: post #6

So openAI's answer to Perplexity's Comet. I'm afraid this will be the future, as these AI-browsers do truly bring value. But they open up the gate for a single Big Tech Winner that truly knows everything about you, and can even control everything on your behalf. I really hope open-source Browsers like Firefox follow up soon with better alternatives, like on-device LLMs to counteract the "all in the cloud" LLM approac…

Websites are terrible from a security, usability, accessibility, privacy, and mental health perspective. These tools could be used to fix all of those things. Instead they're just being used to do the same old junk, but like... faster.

I want an AI browser that digs into webpages, finds the information I want and presents it to me in a single consistent and beautiful UI with all of the hazards removed. Yes, I even want the stupid machine to filter content for me. If I tell it "no politics on Tuesdays" it should be able to go find the things I'm interested in, but remove the references to politics.

I understand that there are new risks to this approach, but it could be built with those things in mind. I'm aware that this would give a lot of power to the developers, but frankly trusting thousands/millions of individual weirdos on the open web hasn't turned out to be any better at this point and it's all become consolidated by near monopolies in user-hostile ways anyhow.

Re: ChatGPT Atlas

#74

Post by webdevver »

behnamoh wrote: post #61
jstummbillig wrote: post #39

Earlier quoted context omitted.

How else would you build new products?

building and shipping are not the same.

building, shipping, and then discarding when it becomes apparent that the product doesn't have a future, is about as good as it gets. most businesses have trouble doing just one, let alone all three.

Re: ChatGPT Atlas

#76

Post by throwup238 »

wild_pointer wrote: post #21

But is it secure? Anthropic demonstrated how agentic browsers are all but. https://www.anthropic.com/news/claude-for-chrome?s=33

Hell no it’s not secure. They brought up the safety issue in the livestream.

There’s a toggle when doing agent mode between “Logged In” and “Not logged in” which I assume uses a temporary chrome profile to wipe all sessions/cookies/local storage for the request. That’s quite a setting for a consumer product.

Re: ChatGPT Atlas

#77

Post by anuramat »

behnamoh wrote: post #25

Anyone feel like OpenAI is acting like Google lately? They announce a lot of products/features and then kill them when they realize people don't use them[0]. They also announce products way before they're ready for launch, just like google[1]. - GPT Plugins? (HN went crazy over this, they called it the "app store moment"...) - GPTs?! - Schedules? - Operator? - The original "codex" model? [0]: I know, the diff is that…

how is codex on this list though? "agent" is also alive for now, and I doubt it's gonna go down any time soon

Re: ChatGPT Atlas

#78

Post by srcreigh »

sansseriff wrote: post #20

I haven't used LLM chrome plugins because I couldn't trust that they weren't collecting more information about my browsing than I'd like. The same concern exists for this, though now I'm just confident it's a giant software company with access to my data rather than some shady plugin developer. I'm faced with asking myself if that's actually better...

OpenAI pins certificates on their macOS chatgpt app, so it’s hard to monitor the data they’re collecting.

Re: ChatGPT Atlas

#79

Post by mentalgear »

echelon wrote: post #42
mentalgear wrote: post #6

So openAI's answer to Perplexity's Comet. I'm afraid this will be the future, as these AI-browsers do truly bring value. But they open up the gate for a single Big Tech Winner that truly knows everything about you, and can even control everything on your behalf. I really hope open-source Browsers like Firefox follow up soon with better alternatives, like on-device LLMs to counteract the "all in the cloud" LLM approac…

If the DeepSeek approach to training hyperscaler models "cheaply" after all the hype wears down works, we just need to follow in their footsteps and build open source alternatives to everything. Frontier models take a lot of money and experimentation. But then people figure out how to train them and knowledge of those models and approaches leaks. Furthermore, we can make informed guesses. But best of all, we can exfi…

Indeed the "no-moat" (weight exfiltration of any openly accessible model) is something that makes me optimistic. That, and the fact that most tasks have "capacity thresholds" which on-device models will increasingly be able to saturate. One example is SQL query generation from text (example: duckdb-text2sql https://motherduck.com/blog/duckdb-text2sql-llm/).

Re: ChatGPT Atlas

#80

Post by SeanAnderson »

jerjerjer wrote: post #17
SeanAnderson wrote: post #4

How does this page not immediately address what I assume is everyone's first question? Is this browser built on Chromium, or is it a completely fresh creation? I have to assume that because they AREN'T highlighting it that it IS built on Chromium.

Why wouldn't it be Chromium/V8-based? New browser engine is multiple developer-years of effort.

Just seems like not a very smart long-term move to build on it.

Tough to get changes upstream when the majority of engineers working on Chromium are Google devs.

Lots of features you'd hope would be available in Chromium aren't there and have to be implemented manually, but then you need to keep your fork interoperating with a massive, moving target. Safe Browsing, Translate, Spellcheck, Autofill, Password Manager aren't available in Chromium and Google cut unauthorized Chromium browsers off from using Google Sync in 2021 (https://www.zdnet.com/article/google-cuts-off-other-chromium...)

There's probably more issues?

Post reply on HN