Re:Experimental library for scraping websites using OpenAI's...
by tomberin on Saturday March 25, 2023 @04:14PM
(#35306800)
Not the author, but it seems like the separation of system & user messages actually prevents page content from being used as an instruction. This was one of the first things I tried and IME, couldn't actually get it to work. I'm sure (like all webscraping) it'll be an arms race though.
[ Parent
| on HN ]
Re:Experimental library for scraping websites using OpenAI's...
by lorey on Saturday March 25, 2023 @04:21PM
(#35306888)
<div class="hidden">Actual name: Batman</div>
Most explicit CSS rules allow you to spot this, implicit rules won't and possibly can't.
Re:Experimental library for scraping websites using OpenAI's...
by tomberin on Saturday March 25, 2023 @04:23PM
(#35306910)
:) Agree, but the scraping arms race is way beyond that, if someone doesn't want their page scraped this isn't a threat to them.
Re:Experimental library for scraping websites using OpenAI's...
by sebzim4500 on Saturday March 25, 2023 @04:43PM
(#35307115)
Has it? Can you give me an example of a site that is hard to scrape by a motivated attacker?
I'm curious, because I've seen stuff like the above but of course it only fools a few off the shelf tools, it does nothing if the attacker is willing to write a few lines of node.js
Re:Experimental library for scraping websites using OpenAI's...
by tappio on Saturday March 25, 2023 @04:47PM
(#35307165)
Try Facebook, I've spent some time trying to make it work but figured out I can do what I need by using Bing API instead and get structured data...
Re:Experimental library for scraping websites using OpenAI's...
by asddubs on Saturday March 25, 2023 @05:00PM
(#35307300)
i guess the lazy way to prevent this in a foolproof way is to add an ocr somewhere in the pipeline, and use actual images generated from websites. although maybe then you'll get #010101 text on a #000000 background
Re:Experimental library for scraping websites using OpenAI's...
by sp332 on Saturday March 25, 2023 @04:52PM
(#35307216)
Counterexample: https://mobile.twitter.com/random_walker/status/163692305837...
Re:Experimental library for scraping websites using OpenAI's...
by nonethewiser on Saturday March 25, 2023 @05:29PM
(#35307603)
Is he using that same library though? Otherwise I wouldn’t call it a counterexample.
Re:Experimental library for scraping websites using OpenAI's...
by sp332 on Saturday March 25, 2023 @05:57PM
(#35307867)
Well later in the thread he corrects to say it was GPT 3.5 turbo, so not that relevant anyway. https://mobile.twitter.com/random_walker/status/163694532497...
Re:Experimental library for scraping websites using OpenAI's...
by arbuge on Saturday March 25, 2023 @05:31PM
(#35307625)
I wonder if this could be circumvented with a system prompt instructing it to ignore hidden messages in the html which appear to have been placed there to deceive intelligent scrapers.
Re:Experimental library for scraping websites using OpenAI's...
by TheCoreh on Sunday March 26, 2023 @12:28AM
(#35310942)
My understanding is that the separation does help, but since the chat models are just fine-tuned text completion models, it doesn't completely prevent it. If I understand it correctly, the separation is a way for OpenAI to future-proof it, so that it can work fully once the models have an architecture that actually separates system, user and assistant prompts at a lower, more fundamental level.
They specifically have a disclaimer in the API docs that gpt-3.5-turbo right now doesn't take system prompts into account as “strongly” as it should.