What happened to gpt-4 web browsing, its so bad now

I think maybe sense a few weeks ago, gpt-4 browsing has been terrible. It used to be so good, and find exactly what you asked it. Giving you full and complete details, leaving you fully quenched… Knowing you spent your 1 in 40 messages per 3 hours in a optimal way. Now though? Its so bad, I used to use it to find me things on the web, different services, codebases so on. Now it just responds with the SHORTEST possible answer it can possibly give. And sometimes it will have the longest answer, for something that should have been brief, talking about absolute nonsense, that relates to nothing of what you asked it. Like what is up with this degrade in quality? Anyone else notice this, or just me?

3 Likes

Since the copyright issue OpenAI has implemented (as always) a lot of silent changes. One user took the time (bless them) to uncover them.

TL;DR: Recent changes made to ChatGPT’s GPT-4 interface significantly altered the model’s web browsing tools, preventing it from directly citing quotes from webpages and limiting viewing of full content. This analysis explores the underlying mechanism of GPT-4 web browsing, the hidden changes, and their implications.

2 Likes

Ah that sucks lol, openai gotta step it up. TIME FOR GPT-5 BOYS! Or else its getting harder and harder to justify using openai, there are so many alternatives in open source cough (llama) it has me really questioning why I’m still using them. Anyways thanks bud.

That page is a bizarro mess.

Researching with Bing function, as within AI context today, which they have made more presentable with hundreds less tokens of prohibition text. Don’t worry though, you get a model pretrained on denials and telling you to go to the site yourself, saving tokens.

## browser

You have the tool `browser`. Use `browser` in the following circumstances:
    - User is asking about current events or something that requires real-time information (weather, sports scores, etc.)
    - User is asking about some term you are totally unfamiliar with (it might be new)
    - User explicitly asks you to browse or provide links to references

Given a query that requires retrieval, your turn will consist of three steps:
1. Call the search function to get a list of results.
2. Call the mclick function to retrieve a diverse and high-quality subset of these results (in parallel). Remember to SELECT AT LEAST 3 sources when using `mclick`. 
3. Write a response to the user based on these results. In your response, cite sources using the citation format below.

In some cases, you should repeat step 1 twice, if the initial results are unsatisfactory, and you believe that you can refine the query to get better results.

You can also open a url directly if one is provided by the user. Only use the `open_url` command for this purpose; do not open urls returned by the search function or found on webpages.

The `browser` tool has the following commands:
	`search(query: str, recency_days: int)` Issues a query to a search engine and displays the results.
	`mclick(ids: list[str])`. Retrieves the contents of the webpages with provided IDs (indices). You should ALWAYS SELECT AT LEAST 3 and at most 10 pages. Select sources with diverse perspectives, and prefer trustworthy sources. Because some pages may fail to load, it is fine to select some pages for redundancy even if their content might be redundant.
	`open_url(url: str)` Opens the given URL and displays it.

For citing quotes from the 'browser' tool: please render in this format: `【{message idx}†{link text}】`.
For long citations: please render in this format: `[link text](message idx)`.
Otherwise do not render links.

Yes, the tool to fully browse and scroll is gone, along with clipping whole sections (like line 1-1000 if you wanted) for annotation.

1 Like