Hacker Newsnew | past | comments | ask | show | jobs | submit | throwaw12's commentslogin

So what?

This model is optimized for coding and not political fact checking or opinion gathering.

If you go that way, with same success you can prove bias in western models.


> with same success you can prove bias in western models.

What are some examples? (curious, as a westerner)

Are there "bias" benchmarks? (I ask, rather than just search, because: bias)


This isn't a result of optimizing things one way or another

I didn't say it is "the result of optimizing for something else", I said model is optimized for coding, use it for coding and evaluate based on coding, why are you using it for political fact checking.

when do we stop this kind of polarization? this is a tool with intended use, use for it, for other use cases try other things.

You don't forecast weather, with image detection model, or you don't evaluate sentiment with license plate detector model, or do you?


> when do we stop this kind of polarization?

When the tool isn't polarized. I wouldn't use a wrench with an objectionable symbol on it.

> You don't forecast weather with image detection model

What do you do with a large language model? I think most people put language in and get language out. Plenty of people are going to look askance at statements like "the devil is really good at coding, so let's use him for that only". Do you think it should be illegal/not allowed to not hire a person because they have political beliefs you don't like?


Neither is the bias and censorship exhibited in models from Western labs. The point is that this evaluation is pointless. If it's mission critical for you to have that specific fact available to the model then there are multiple ways to augment or ablate this knowledge gap/refusal.

This is super cool. Of course, it is possible to separate instrument sounds using specialized tools, but can't wait to see how people use this model for bunch of other use cases, where its not trivial to use those specialized tools:

* remove background noise of tech products, but keep the nature

* isolate the voice of a single person and feed into STT model to improve accuracy

* isolating sound of events in games and many more


easier said than done.

Parents had enough problems to think about.

In a similar way we can say that every shop in Amazon can create own digital shop themselves, but marketing, sales channels and distribution is not easy to acquire.


I have reported these ads to YouTube multiple times, because I tracked down their scam websites, but YouTube didn't delete them anyway.

Common pattern they had was:

- similar or same domains

- same messaging on their website

YouTube could have taken action, but it choose not to


I'm still waiting for the tech world to wake up and realise that the online ad machinery and user tracking software that the brightest minds of our generation have been working on are just a way to efficiently connect scammers with their unsuspecting victims.

Oh, they know that. It's very lucrative. At this point it's scams all the way up to the US presidential cryptocurrency.

However it's also a tricky business to be the adjudicator of what is and isn't a scam. You're going to have to deal with a lot of complaints from "legitimate businessmen".


I'm waiting for the non-tech world to wake up and hold companies that act as willing accomplices liable for the crimes they tolerate on their platforms.

> the crimes they tolerate on their platforms.

... the crimes they actually make a lot of money from.


The tech world knows this. They are raking in money off of these scams. People with a rudimentary moral compass leave, those without stay, which makes it even less likely that industry will self-sanitize. The rest of society, out of survival instinct if nothing else, will have to force it to stop anti-social and fraudulent practices. Same as many other industries.

I'm waiting for the tech world to realize that "the brightest minds of our generation" don't actually work at google, because if you are that enormously bright you don't want to work for ads or in an opaque megacorp.

Why does anyone think a brilliant mind would enjoy that? So they could make a little bit more money?

Do you honestly think brilliant people, the smartest of our generation, care about money?

IME, Google software devs aren't even the brightest minds in the parking lot.

Completing large engineering projects says nothing about individual capability, and nothing about how Google deploys shitty AI moderation and about how Google employees insist it's great and perfect and never does anything wrong gives me any reason to believe they are even competent.

It's literally a meme that people started repeating in earnest without a second thought.

Don't you think a brilliant person would work somewhere, like, interesting?

In economies where you aren't rewarded for individual competency (because software management couldn't pick out individual competency if it screamed at them), highly competent people aren't going to play the game, they are just going to find something to pay the bills and work on hobbies.

The smart people are often where the money isn't, because they are rarely driven by monetary pursuits.


It would help to stop saying "brightest minds of our generation", like we stopped saying "smartest guys in the room".

They are not the brightest, just the ones who sold out others and grabbed the money, with ethics and morals not being sufficient personal barriers.

Calling them the brightest just feeds their belief that they merit the money, and they don't have to ask the real reason they have so much money.


They already know. Meta estimates around 10% of its ad revenue comes from ads for scams or banned goods. https://www.reuters.com/investigations/meta-is-earning-fortu...

I think being a “Techie” is now something that is splitting.

- People who want to work in tech because it was a stable and/or lucrative career

- People who just want/love to code

- People who loved tech / think tech is cool

There’s also a degree of counter-culture that used to be part of the mix, which got jettisoned as tech became mainstream and mapped out.

The current state of Tech is unpleasant and alarming.


It is difficult to get a man to understand something when his salary depends on his not understanding it.

(but enough about LLM shills)

There was a period when I was constantly showered with these ads whenever I visited YouTube. It quickly became clear that it was some kind of scam, but YouTube didn't do anything about it for years.

Does clicking on the ad cost the spammer a lot of money

Yes, which is one of two reasons why I use a blocker called adnauseum. It an adv locker that “clicks” on every single ad it sees, as well as hides it from my view. This makes my ad profile useless, and also costs them money.

you should probably think about the fact that ad platforms expect and design for fraudulent and bot clicks before you assume that this actually costs anyone money.

The question is, is the scammer taking donations from kids with cancer or Google the more worthy entity to profit from the situation? It's a tough decision.

> This makes my ad profile useless

I have a theory that it doesn't. Which set of companies' logic is more likely?:

Is LadyCailin a "tree-hugging liberal"? LadyCailin clicked on a lot of Sierra Club and PETA ads, so yes. Good, we will add LadyCailin to this list.

Is LadyCailin an "extremist right-wing nazi"? LadyCailin clicked on a lot of prepper and gold ads, so yes. Good, we will *also* add LadyCailin to this other list.

OR

Is LadyCailin a "tree-hugging liberal"? Well, they clicked on these ads, so we think so, but then they clicked on these other ads, so we're not sure. Then she clicked on these other ads, now we don't have any idea.

Speaking from personal experience: Because some people have used my phone number and email address as their own, I get emails for one political party and text messages for the other political party.

It doesn't make my ad profile useless to the people sending me ads.

Give your phone number to both U.S. political parties. Congratulations you will get spammed by both. I doubt they are cross-checking.


Won’t Google ban your account if they notice this

As per other comments, if it’s making them money, why bother banning it

Depends on bot-detection and if its a CPC or CPA campaign

Once you realize how profitable it is, it's hard to stop. https://www.cnbc.com/2025/11/06/meta-reportedly-projected-10...

see Meta, which is operating like a crime syndicate, leveraging higher fees on scammers "to discourage" them, retaining their impact on supply side auction prices and well knowing many don't pay with their own credit cards.

https://www.reuters.com/investigations/meta-tolerates-rampan...


They didn't care about being a huge driver behind a genocide, why would they possibly care about people getting scammed out of cancer fundraiser donations. Once you find out that someone has worked at any point for Meta past say, 2020, you know everything you need about them.

[1] https://www.amnesty.org/en/latest/news/2022/09/myanmar-faceb...


Same. Even if they delete one it's usually delayed for 2-3 days. The worst part about scam ads is that they surface a day later from a different account with 0 changes to the ads themselves. You would think Google would fingerprint the assets but in the end they just don't care.

There's no incentive for them to comply with your request. Like Facebook, scam ads are a revenue stream for Google. The profitability usually offsets any negative PR or fallout that results from these platforms turning a blind eye to the point where their budget accounts for some percentage of scam income, leaving them to pick and choose when to take action while they actively make their platform increasingly hostile to users who want to protect themselves from said ads.

Google, the worlds biggest and best coconspiritor.

That's so Meta.

They also had a pattern of loudly crying kids in the beginning of the video, I thought they were faking, after a month they changed the style of start.

Scams are extremely high margin businesses and as such can spend very generously on advertising. Consequently the Googles of our world love them.

In my experience, anything related to Google Ads - they never reacts to any claims of scam…

Their incentives contradict healthy behavior… :(


Yep. Lately I've been getting dozens of scam ads for pulse oximeters being sold as Glucose meters, with a big ol' FDA logo plastered over the top of the video. A flagrant violation of regulations around medical device marketing.

Here's Google's response:

  We understand you are concerned about the content in question, but please note that Google's services host third-party content. Google is not a creator or mediator of that content. We encourage you to resolve any disputes directly with the individual who posted the content.
...which is a lie, among other things.

What struck me is that when I reported an ad with an Elon Musk deepfake selling some crypto scam, I got an email back from Google saying that after reviewing the video they found nothing wrong. I don't understand how this is not actionable in court- I mean, you did act on a report, you declare you manually reviewed the content and that it's good for you? I don't get it.

What's most depressing is that people like you did the right thing (took the time to investigate and report) and still hit a wall

reminder that according to Facebook's own analysis 10% of their 2024 revenue comes from scams and banned products

"Is it possible to get a fair and just trial in China?"

What makes you ask a such question? Here are some bad ideas which comes to my mind:

* you think China is inferior?

* or maybe Chinese are inferior?

* maybe you think they always lie?

* or maybe they don't have laws?

* maybe plain old racism?

Forgive me, but your question sounds so bad. Counter question, did any of war criminals get a fair trial in the USA? (I am not listing countries they did war crimes, because there are too many)


> What makes you ask a such question?

Obviously, it is that a political opponent of the administration is facing life in prison seemingly for being an outspoken critic of the administration.


Calling on other nations to harm your country is generally considered treason in any country.

If you own a newspaper and use that platform to make such a request, you're likely to attract a law enforcement response wherever you are.


He isn't a political opponent, he isn't a politician. He met with high-ranking officials of a foreign nation and lobbied them to take action that he believed would harm China and lead to a change in government.

I also wouldn't call him outspoken critic either. For obvious reasons, the main one being a level of economic development unknown in human history, there isn't very much to criticize outside of politics. His gripe is solely political in that he believes that a different system of government is required (one assumes with more input from people like himself, again though he isn't a politician and, afaik, has no real political positions apart from supporting Trump and NY Post-style sensationalism/xenophobia, iirc they created a meme depicting mainlanders as locusts...it is quite funny to see people who, I can only assume, are not massive fans of Trump cream themselves over the Chinese equivalent).


Is the US currently running massive concentration camps? Is the US currently involved in large scale organ harvesting?

> Is the US currently running massive concentration camps?

https://www.ice.gov/detention-facilities


China isn't running massive concentration camps, nor is it involved in large scale organ harvesting.

"Democracy that we knew from 60s to 00s is effectively dead everywhere."

What if it was a fake all along and was just a facade and social media exposed it?


It was, but what we have no is also much worse. Both can be true.

Who is the beneficiary from this?

I don't think its EU citizens, because:

    * roads will be damaged faster
    * risk of hitting and killing more people
    * because roads damaged more tax money spent on fixing them
    * more CO2
I think EU should go back to build good relationships with Russia, take its cheaper gas & energy and support its own economy, instead of propping up the US economy and opening the market for its ugly huge cars.

Just come to Amsterdam and see if you can drive those cars in the middle of Amsterdam. Even trams from 2 opposite direction share same line in some areas.


The EU was bending over backwards for Russia until they invaded a neighboring country for being too friendly to them. The fact that relationships aren't good there is entirely on Russia.


Build good relationships with Russia? That's a call to Russia, not the EU! First and foremost, Russia has to stop going to war with its neighbours. In any case, Europe doesn't need Russian gas for much longer.


You think EU should go back to building good relationships with Russia when there is an ongoing war of aggression started by them? If you really believe that and you're an EU citizen I can't help viewing you as traitorous to very foundational values the EU was created for. Absolutely disgusting.


[flagged]


Every point you mentioned is just parroting the Russian narrative.

There's no such thing as "good trade relationships with Russia", as those that were there existed only thanks to planted agents like Gerhard Schröder.

What most likely triggered this war was Putin's ambition to stay in power, as Russia never actually recovered from the 2008 crisis, so he let Medvedev handle the popularity hit associated with the first years post that.

Russian agents are sabotaging European businesses as we speak - there's no getting back to whatever level of friendly relations there were before the invasion.


Gerhard Schröder was no planted agent, he acted out of a belief in what he did. The german left have always had a strong connection to russia.

I live in northeast Germany and the cost of living increase and industrial cost of the nord stream & oil pipeline changeover has been immense.

I am not advocating a "friendly" relationship w russia but it's also wrong to over simplify the relationships. The nord stream sabotage was a HUGE gut punch for former east german states.


And yet NATO expands most readily when Russia invades another country. I wonder why nations might want in that alliance? You're just repeating Russia's justifications for their actions, which have never made sense.


NATO expansion was because countries begged to become members of NATO, out of fear of Russia invading them. The only reason Russia minds NATO expansion is that it prevents them from starting easy wars.


You forgot the bio weapon labs! /s


> I think EU should go back to build good relationships with Russia

Kinda hard with someone trying to expand, starting wars and engaging with genocide. Literally.

Being accommodating to Russia is how we got here.


> I think EU should go back to build good relationships with Russia

This is horribly naïve at best. You're suggesting building good relationships with a country waging a war of aggression with a neighbour it shares with the EU. A country that's committing genocide against that neighbour. A country that has been rather consistently stepping up its attacks against European infrastructure over the past several years.

I'm not saying that you are an idiot. But I am saying that you would have to be an idiot to sincerely believe what you just said.


What happens if Voyager decides to go back home?


I am curious as well about this, we use Snowflake, but as a software engineer I want to understand how Spark/Databricks is different, what are we missing out?

How we work with data is simple, if SQL+dashboard solves the problem then we do it in Snowflake, if we need something more advanced, then code + bunch of SQL.

Pretty sure ML engineers work in different ways, but I don't know that side well


I would pose a question differently, under his leadership did Meta achieve good outcome?

If the answer is yes, then better to keep him, because he has already proved himself and you can win in the long-term. With Meta's pockets, you can always create a new department specifically for short-term projects.

If the answer is no, then nothing to discuss here.


Meta did exactly that, kept him but reduced his scope. Did the broader research community benefit from his research? Absolutely. But did Meta achieve a good outcome? Probably not.

If you follow LeCun on social media, you can see that the way FAIR’s results are assessed is very narrow-minded and still follows the academic mindset. He mentioned that his research is evaluated by: "Research evaluation is a difficult task because the product impact may occur years (sometimes decades) after the work. For that reason, evaluation must often rely on the collective opinion of the research community through proxies such as publications, citations, invited talks, awards, etc."

But as an industry researcher, he should know how his research fits with the company vision and be able to assess that easily. If the company's vision is to be the leader in AI, then as of now, he seems to have failed that objective, even though he has been at Meta for more than 10 years.


Also he always sounds like "I know this will not work". Dude are you a researcher? You're supposed to experiment and follow the results. That's what separates you from oracles and freaking philosophers or whatever.


Philosophers are usually more aware of their not knowing than you seem to give them credit for. (And oracles are famously vague, too).


Do you know that all formally trained researchers have Doctor of Philosophy or PhD to their name? [1]

[1] Doctor of Philosophy:

https://en.wikipedia.org/wiki/Doctor_of_Philosophy


If academia is in question, then so are their titles. When I see "PhD", I read "we decided that he was at least good enough for the cause" PhD, or PhD (he fulfilled the criteria).


he probably predicted the asymptote everyone is approaching right now


So did I after trying llama/Meta AI


He's speaking to the entire feedforward Transformer-based paradigm. He sees little point in continuing to try to squeeze more blood out of that stone and instead move on to more appropriate ways to model ontologies per se rather than the crude-for-what-we-use-them-for embedding-based methods that are popular today.

I really resonate with his view due to my background in physics and information theory. I for one welcome his new experimentation in other realms while so many still hack away at their LLMs in pursuit of SOTA benchmarks.


If the LLM hype doesn't cool down fast, we're probably looking at another AI winter. Appears to me like he's just trying to ensure he'll have funding for chasing the global maximum going forward.


> If the LLM hype doesn't cool down fast, we're probably looking at another AI winter.

Is the real bubble ignorance? Maybe you'll cool down but the rest of the world? There will just be more DeepSeek and more advances until the US loses its standing.


How is it a foregone conclusion that squeezing the stone will continue to produce blood?


I believe that the fact that Chinese models are beating the crap of of Llama means it's a huge no.


Why? The Chinese are very capable. Most DL papers have at least one Chinese name on it. That doesn't mean they are Chinese but it's telling.


is an american model chinese because chinese people were in the team?


There is no need for that tone here.


OP edited the post.


What are these chinese labs made of?


500 remote indian workers (/s)


most papers are also written in the same language, what's your point?


LeCun was always part of FAIR, doing research, not part of the LLM/product group, who reported to someone else.


Wasn't the original LLaMA developed by FAIR Paris?


I hadn't heard that, but he was heavily involved in a cancelled project called Galactica that was an LLM for scientific knowledge.


Yeah that stuff generated embarrassingly wrong scientific 'facts' and citations.

That kind of hallucination is somewhat acceptable for something marketed as a chatbot, less so for an assistant helping you with scientific knowledge and research.


I thought it was weird at the time how much hate Galactica got for its hallucinations compared to hallucinations of competing models. I get your point and it partially explains things. But it's not a fully satisfying explanation.


I guess another aspect is - being too early is not too different from being wrong.


then we should ask: will Meta come close enough to the fulfillment of the promises made, or will it keep achieving good enough outcomes?


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

Search: