Hacker Newsnew | past | comments | ask | show | jobs | submit | neksn's commentslogin

> We do not have a say over a nuclear-armed idiocracy forcing it's profound corruption and stupidity on other sovereign entities.

That’s funny because it perfectly defines the relationship between the EU and the countries in it.


Interestingly, using AI Studio with the gemini-2.5-flash-image model and asking it to generate a "completely black image" fails with the message "Image recitation block," which has zero hits in Google.


"image recitation block" means they are blocking generation of images that already exist in their database (training data).


Just to confirm, what this means is that the training data contains one or multiple pitch black images, and their system is refusing to reproduce them verbatim?


no insider knowledge here. my assumption is that the image hash matches a training data image. "all black" is a pretty easy hash to match.


Considering all models can use search engines, is this really relevant?


Yes. Huge difference in quality in from-weights distilled knowledge vs something based on a search tool. If the LLM uses a search tool there's barely a difference between a 30B model and Opus or GPT 5.5, because it just bases its reply on the stuff that came up. Which is generally SEO junk.

Obviously with the last example I'm not talking about long-running agentic tasks here that involve many dozens of search calls (like the recent Erdos problem stuff).

And that doesn't even consider the extra content rot, the time it takes, the need for such an API and so on.

One of the biggest advantages Anthropic models have had over GPT was GPT's woefully outdated data cutoff. They finally improved on this with 5.5, but IIRC it took a year.


This is not meant as an insult, but have you actually LLM/vibe coded anything that used a fast(-ish) moving library or framework? Try asking your favorite LLM with say Jan 2025 knowledge cutoff (or pretraining data cutoff, whatever you want to call it) to work on something using a framework that had a big rewrite later that year (which would make it one year old now, which is like ages in the LLM coding era)... It's a nightmare full of wrestling with the LLM when you try to tell it the version of the framework and that it changed a lot from the previous version and yadda yadda long story short down the thread when context runs out and/or is compressed it begins to forget detailed instructions and just falls back to pulling out old patterns it "remembers" from pretraining. And so you need to constantly remind it what you work with and "oh hey this doesnt work because we're working with react router v7 in framework mode, remember? not react router v6". Or try to use the latest non-lts/breaking version of a library, at first it looks it up online, but again as you get deeper into the weeds and little details, the struggle begins.

So, as far as I'm concerned, training cutoff is still a big deal.


> It's a nightmare full of wrestling with the LLM when you try to tell it the version of the framework and that it changed a lot from the previous version and yadda yadda

Tip: Add a default instruction to look at the actial downloaded source code of the dependencies used (assuming you're not dealing with closed source dependencies). Have the agent treat it as your own (readonly) source code instead of relying on model training data and possibly mismatching documentation on the web. Then it just greps for the exact function signatures and reads the file based documentation.


Great, now you experience context bloat 3x as quickly and any task takes 3x as long.

Ifz Google wants to structurally compete with Anthropic on coding, this issue is a must-fix. OpenAI finally fixed it with 5.5.


Until they prefer not to search. Let me explain using the example of the open-source security framework (1) our team is working on.

If you ask Gemini what you should use to integrate fraud prevention or account takeover protection into your product, there will be no mention of our open-source project. Five years in development, 1.3k stars, over 140 pull requests — all this isn't enough to make it into the training data. From this perspective, any technology that emerges after 2024 is simply invisible to LLMs.

The answer is: without being in the training data, LLMs basically don't understand what they're searching for.

1. https://github.com/tirrenotechnologies/tirreno


I just put the terribly generic query "what tools would you recommend to integrate fraud prevention or account takeover protection into my product" into both Claude (Sonnet) and Gemini (3.1 Pro) via the standard web interface and both took the first step of searching the web. That's consistent with my past experience -- the usual harnesses typically will search the web in cases where I might expect/want them to. Now whether you product has good web visibility or not in those searches and how the LLM's weigh the relative merits of open-source tools versus commercial offerings in deciding what to highlight in their responses is a different issue. As is the change in what constitutes effective SEO in an era where bots, rather then human eyes are the proximal important target. But I don't think the core issue with folks finding your products is the move away from user-driven search toward using models with out-of-date training cutoffs.

FWIW while neither model included your product in it's initial response, when I followed up with "what about open-source" both did another search and Claude's response included your tool....


> while neither model included your product in it's initial response, when I followed up with "what about open-source"

You just proved that LLMs don't know about the product (which is fine), but they don't even know the category exists.

It's like driving a car whose mirrors show a two-year-old reflection and insisting they work fine.


> And absolutely no one seems to be interested in answering the question of “okay, then what?”

I don’t see why the people being booed should be responsible for answering this question. How many such questions did the inventor of the tractor have to answer?


The tractor displaced horse and oxen.

Which were slaughtered when no longer needed.

You should rethink your metaphor because it's not having the effect you intended.


Imagine if the inventor of the tractor went to a college for farm workers (if there were such a thing) and gave a commencement speech that was all, "Tractors are going to revolutionize farming by making your jobs obsolete." I think it would be fair to expect some answers about how the new graduates should handle that. Or maybe Mr. Tractor should just stay home if he doesn't have the answers or doesn't want to face the crowd.

This isn't "people are upset with AI and demanding answers from the people creating it." This is, "the creators are showing up at schools and giving speeches about how everyone is fucked, and this is getting a bad reaction for some unfathomable reason."


They caused the problems, so they are responsible.


It would absolutely have been valid to ask that question of the inventor of the tractor too.

It's even more relevant to ask of the CEO/CTO/COO/etc. of the companies that are selling hard on eliminating humans from as many workflows as possible.


They are selling a reduction in labour costs which has been the primary selling point of automation since humans began automating things.


Yep. They are talking to other CEOs, not to the young graduates they are supposed to be talking to.


I can’t even imagine how ridiculous this will look.

How about you go back to making actual cars at that price point?


Looks a lot like a Fiat Panda, e.g. an "actual car": https://www.lautomobile.aci.it/attualita/stellantis-arriva-l...


I was thinking more in terms of horsepower, etc. I’m sure it will be a “city car” with ridiculous power and autonomy.


The page is gzipped in transit - only 5 MB of traffic are generated.


Isn’t that the audio description (AD) track? Maybe it’s mixed in because of an encoding error.



I wish... it is much more generic and touches on the plot which the audio description track wouldn't do. I'll see if I can find one. They are extremely annoying.


…she was told she had to pay taxes that she didn’t owe. It was a shakedown, which is quite common for the Spanish tax agency.


And I understand it when you owe tens of millions of euros and plenty of resources to make everything properly.


She did everything properly, but the tax agency illegally imposed a fine on her anyway.

Remember that in Spain the tax agency loses 51 % of trials. I assume the percent would be even higher if everybody could afford to defend themselves.


And the other 20 million Euros she settled for the next fiscal years?


A Mandela effect is when it happens to a large group of people. I’m afraid you’re alone in this.


Only when large group of people die and have to be switched to another timeline via quantum immortality.

This person died independently. Welcome to our timeline!


What’s interesting to me is that for you to build a $10B datacenter (or any other business) you are shaken down for over $3.3B.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: