This article appears to be a coordinated attack on Perplexity.

Also posted today is this HN thread:

https://news.ycombinator.com/item?id=49536375

Containing this link:

https://trellner.com/reports/manufactured-sources-behind-ai-...

Which has the exact same layout and very similar About page as "trellner.com"

https://hausresearch.com/about/

https://trellner.com/about/

Same is true of Google search “summaries” where quite often I click on the link provided and it doesn’t support the statement Google made.

BTW, this post would be more convincing if it wasn’t written in Claude voice itself!

There’s a fun variation in W-Europe that google needs to spend some time on:

Northern Belgium and the Netherlands have web content in the same language. But google uses the content in one lump. Problem is when you search for employment/fiscal/legal/… you constantly get content that applies to the wrong nationality.

Fits studies on multilingual IR generally: relevance ranking treats language as proxy for locale, which breaks down whenever two jurisdictions share a language. Same failure mode shows up in legal databases indexing US and UK case law together.

This is true of a ton of online discourse. Worse, when the headline of a claim doesn't even match the article it is fronting. I've seen more than a few articles that basically contradict the headline, but end in a "despite all evidence, we think it is correct to say X."

Never believe the citations in Google summaries, I can count on a single hand the time I have found a correct citation.

It's not 2/3 correct, it's <1% correct, in my experience. Maybe it's because I check things that sound off more frequently, but even random checks have not panned out well.

Anybody relying on Google AI summaries is misinforming themselves.

<1% seems steep, but IIRC Google's summaries pull snippets not full citations, so "correct" is fuzzier to define. Doesn't change your point though: if you actually check them, they fall apart constantly.

Those summaries (and Bing is at least as culpable) are mostly only good for comedy value. Most favourite genre: returns picture of someone you know alongside biography of completely unrelated person. Least favourite genre: useful stat supposedly cited from a linked reliable source (but turns out to actually be a different stat in the reliable source)

I don’t know if Claude performs similarly from a percentage standpoint, but if you’re using it for search (online or personal docs or wikis), it often also just makes things up.

When you point it out, it’ll do the “ohh you’re absolutely right!” bs. Marketing material and management that believes the material wants to pretend that AI agents are junior employees, but forget that junior employees get fired for doing something like this.

It was true even before AI summaries.

The search result showed a paragraph containing the searched word, the website did not

I don't even know if I disagree with this post, but this seems really astroturfed. Why are there two anti-Perplexity articles from independent research firms with identical websites on the front-page of HN right now, submitted by the same person? Am I going crazy?

(see https://news.ycombinator.com/item?id=49536375, left a comment there also)

This is why lawyers have been getting in trouble using AI to review case law or (worse) to generate documents.

It creates citations and references that look close enough to be plausible but are just made up of thin air. CA passed a law explicitly requiring lawyers to review AI-generated documents that is now before the governor for signing (previously, lawyers were ethically expected to review documents submitted to the court or provided to clients but that doesn't have the same level of force as an explicit requirement).

Even worse than lawyers is the government doing it: https://newrepublic.com/post/215001/judge-rfk-jr-hhs-fake-ai...

Law now requires reading what machine wrote. Bar set low, still tripping over it.

If you are building your own harness that does correct citations, is the correct thing to give AI access to some deterministic tool that allows it to actually copy paste parts of documents its reading (with links), rather than stochastic reproduction that they do by default?

I did something similar for structured text extraction. I added markers throughout each source document and then, for each piece of info I wanted, I asked the LLM to provide two separate fields:

  xyz
  xyz_citation

The latter was just the node number. So then my code could extract the exact snippet, instead of trusting the LLM to quote something verbatim.

That’s what I did when I built my stuff. I have deterministic content with AI commentary, where it seems most people are doing this crazy thing of sending data through the model. I can’t understand it.

You can just look at nouswise or nblm. They do have such harness behind.

I noticed this personally. Saw a citation with a preview for source A, which I knew was reliable. Checked, and it referenced a Reddit article and various other less reliable sources. Was a direct citation too that actually wasn't.

100% my experience with this service. The intent is good, but it seems they're still in the "fake it until you make it" stage.

>The intent is good, but it seems they're still in the "fake it until you make it" stage

Please notice the internal contradictions here.

To be honest perplexity does nothing to make sure it's answer are correct let alone the citations. They just look plausible. For anything little bit serious I use nouswise or nblm that sometimes abstain instead of making things up.

> the intent is good

What is that supposed to mean? They're trying to be an llm search engine that's not some radical new concept

Nobody pages for wrong citations though. That's the tell — no on-call, no SLA, no incident review. Ship it, move fast, let users find the bugs in prod.

Unsurprising, if you’ve ever clicked some of these citations they always seemed like they’re from the back alley of the internet

Now do OpenEvidence :)

I felt that myself. And I have the same problem with gemini

This plus AI just citing AI slop. Theres a real downward spiral unfolding with the quality of information available on the internet.

That’s what I feel too. From the references in their PDF, a few links are returning 404s, while others don’t even seem relevant to the so-called “research.” It is pure AI slop. https://hausresearch.com/data/perplexity-citation-audit/perp...

BS machine produces BS; in other news water is wet and sky is blue.

What's surprising about this is that you can get the bullshit machine to produce correct externally validate citations. It's not particularly hard either—it's one of the first things you build when you give an LLM access to a body of documents/search. So for a large public service to whiff like this is certainly a stain on their credibility.

My favorite is hallucinated slop with sources that 404.

That's the one that powers Truth Social

Trump Media and Technology Group announced that it partnered with Perplexity to test and integrate an AI search feature, referred to as Truth Social AI or Truth Search AI, directly into the Truth Social platform.

(I just use the free account from truth+ to waste their money)

how do you reverse a linked list in python

Answers Sources Use either an iterative pointer-reversal approach or a recursive approach. The standard iterative version is the most common and runs in (O(n)) time with (O(1)) extra space:

class ListNode: def __init__(self, val=0, next=None): self.val = val self.next = next

def reverse_list(head): prev = None curr = head

    while curr:
        nxt = curr.next
        curr.next = prev
        prev = curr
        curr = nxt

    return prev

If you already have a Python list, reversing it is simpler with slicing: items[::-1], but that is not a linked list reversal.

"The unit above is the citation, not the claim." The RPM of the slop ouroboros ever rises.

Ran into this building a RAG pipeline last year. Fix was forcing model to output exact quoted span alongside claim, then string-match it against source before showing citation. Anything that doesn't match verbatim gets dropped, not shown. Cuts hallucinated cites hard.