TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

ChatGPT is hallucinating fake links to its news partners' biggest investigations

50 点作者 giuliomagnifico11 个月前

13 条评论

sixhobbits11 个月前
I like a bit of ChatGPT bashing, same as the next guy, but this seems a bit unfair. The partnerships have been signed, but as far as I know OpenAI has made no indication that the implementation is done or even started yet, so this story is basically &quot;ChatGPT hallucinates URLs&quot;, which is a pretty well-known issue.<p>The &#x27;of partner websites&#x27; bit seems a bit shoe horned in to try prove some point. It hallucinates URLs for any site, and I don&#x27;t think the partnerships are relevant here.
评论 #40821865 未加载
barbariangrunge11 个月前
I think the bigger issue is trust. The chat it is no longer trying to return objective information. “OpenAI” is letting companies pay to have it promote their content—is that accurate?
评论 #40821830 未加载
评论 #40821657 未加载
评论 #40821766 未加载
netruk4411 个月前
I wonder how difficult it would be to make a website that, instead of serving 404&#x27;s, a model does a semantic search using the nonexistent URL to come up with a URL that actually does exist and most closely matches the &quot;intent&quot; of the invalid one.
评论 #40821649 未加载
评论 #40821566 未加载
Kiro11 个月前
I very seldom experience hallucinations and I&#x27;m a really heavy user of ChatGPT. Can someone give me a prompt that is likely to generate hallucinations?
评论 #40822072 未加载
评论 #40828874 未加载
评论 #40822107 未加载
评论 #40822603 未加载
评论 #40821984 未加载
评论 #40822153 未加载
评论 #40821901 未加载
redeux11 个月前
Clearly those companies should use ChatGPT to generate those URLs. That way when ChatGPT guesses what the URL should be, it’ll be correct!<p>&#x2F;s in case it’s not obvious
评论 #40821892 未加载
评论 #40822080 未加载
评论 #40821349 未加载
评论 #40821975 未加载
评论 #40821708 未加载
chrisjj11 个月前
&gt; ChatGPT is hallucinating fake links<p>There&#x27;s no &quot;hallicinating&quot; and no &quot;faking&quot;. A computer program is generating faulty links, period.
评论 #40821853 未加载
Edmond11 个月前
This seems a result of relying on the LLM to accurately extract information that needs to be exact.<p>I touch on this from my own experience: <a href="https:&#x2F;&#x2F;youtu.be&#x2F;cs5cbxDClbM?si=IQIFAD38cVzLCs55&amp;t=486" rel="nofollow">https:&#x2F;&#x2F;youtu.be&#x2F;cs5cbxDClbM?si=IQIFAD38cVzLCs55&amp;t=486</a><p>Basically if you have the actual &quot;factual&quot; information, use it directly instead of hoping the LLM will accurately extract it and use it as part of a function call. In this case they already know what the accurate URLs are, just use it.
评论 #40822001 未加载
maxbaines11 个月前
It&#x27;s easy to think of the models in this context, rather than the application built on top of the model, in this case ChatGPT.<p>ChatGPT is not simply the gpt-4o model, it&#x27;s the model, a system prompt, tools and a virtual environment running python.<p>I built my own app, when using it i don&#x27;t get links to partners because i don&#x27;t mention such a thing in the system prompt.
klyrs11 个月前
It&#x27;s a special kind of hell, taking a hallucinogen and finding yourself in a white buttondown and slacks, writing ad copy.
1vuio0pswjnm711 个月前
Perhaps ChatGPT&#x27;s inherent flaws will drive www users to use newspaper websites instead. This could be a win for journalism, which seems to be the adversary (not the partner^1) of Silicon Valley.<p>1. Except for all the &quot;tech&quot;-columnists pumping out marketing gibberish<p>Assuming the lawyers for the newspapers were smart enough to retain the right to serve the news without &quot;AI&quot;, i.e., they did not agree to funnel users to ChatGPT via their own websites, they become the only authoritative sources, what with www search engine indexes no longer publicly searchable, queries for specific resources having been replaced with &quot;prompts&quot; to word soup generators.<p>With the absurd limits on number or search results and now this &quot;AI&quot; nonsense, I&#x27;ve been preparing to transition away from the popular www search engines toward searching only websites of selected publishers, i.e., authoritative sources. Cut out the middleman. I do all searching from the command line and create mixed SERPs over time, similar to the &quot;metasearch&quot; concept but using site-specific search for non-commercial content instead of www search engines. It has been working well for me.
LetsGetTechnicl11 个月前
LLM&#x27;s?? Hallucinating? Color me shocked!!
评论 #40824866 未加载
localfirst11 个月前
We seem to be exiting the hype phase
curtis338911 个月前
It drives me nuts how OpenAI has misleadingly marketed their products, and the tech community at large has failed to clarify what their products really are. This article is so unsurprising to the point of being boring. Of course an LLM generates broken links. Why wouldn&#x27;t it?
评论 #40821723 未加载
评论 #40821449 未加载
评论 #40821944 未加载
评论 #40821571 未加载
评论 #40822012 未加载
评论 #40821308 未加载
评论 #40821886 未加载
评论 #40821629 未加载
评论 #40822025 未加载