For the best experience on desktop, install the Chrome extension to track your reading on news.ycombinator.com
Hacker Newsnew | past | comments | ask | show | jobs | submit | history | bko's commentsregister

> "I'm writing my code and reviewing my code by hand, which seems crazy to say," said the 34-year-old, who lives in North Carolina and works for a large tech-entertainment company that she described as progressive. "Just two years ago, how else would you do it?"

Imagine working with this person. At a certain point you just can't do the job. What if a religious Jew wants to work at a butcher shop and refuses the handle pork. Why does everyone feel like they're entitled to do a particular job, making everyone else's life considerably more difficult


It’s kinda worse than that - an Orthodox Jew would, by long-standing traditions and teachings, not be expected to touch pork. The same beliefs that make this person object to using AI could easily apply to any cloud-scale service dependent on data centers.

Like say… Google search (the old non-AI one)


Not a good comparison because so many differences between both stories. Also a practicing Jew or Muslim would not work in that kind of butchery serving non kosher meat in the first place, you're trying to delude other people.

???? You can code without ai bud. Leave the frummies out of this. What does kashrut have to do with ai.

Imagine you ask someone "Can you hand me the butter please? That would be very kind." and they respond "Why does everyone in the world want to have all the butter that ever existed in this very instant?!", and then you ask them what they do for a living while chewing on your dry bread, and they say oh, I write software for pacemakers.

> Hallucinations and poor guidance are still a regular day-to-day issue that makes it impossible for me to trust agents with anything.

I often hear this. Can you give me a question where a major LLM hallucinates or provides poor guidance? Reproducible would be great

Just a question to stump it.


Just today, the LLM based auto-review that my company enabled for all PRs edited my PR description to confidently assert that I had added a new RPC. I had not. I deleted code and nothing else. Nothing was added. The RPC it claimed I added did not exist.

This is a common occurrence.


LLMs are nondeterministic, so it’s impossible to make something 100% reproducible. Even if it has an issue, it might do it in a different way. If it’s well publicized, they’ll patch that very specific example, but the foundational issue is still there (like counting the R’s in strawberry).

I still regularly run into the issue where it just makes up API endpoints, CLI commands, or add flags that simply don’t exist.

I also regularly ask it things and it gives me a bad answers, so I push back, and it says something to the effect of “you’re right, I didn’t consider that, let me look at that more”… then tells me the exact opposite of the previous response.

Or it “thing X has never happened”, and I ask what about <insert example>, and it goes to look it up and says, “oh, thing X actually did happen.”

I run into this daily. Multiple times per day. How can I trust a system like this? Are people just blindly accepting what the LLM says as truth? Is that why people think it’s good?


Bad example but since it literally just happened a few hours ago:

Teams Copilot meeting assistant auto-renamed a meeting title/summary that’s now prominently placed at the top to “Month end close wrap up discussion“ because someone posted in chat “sorry can’t make the meeting, we’re wrapping up month end close”.

Really confused the next guy who joined the meeting and derailed things for a minute or two before we could get back on topic.


> Reproducible would be great

Wouldn’t it be great? I’m still waiting for reproducibility from LLMs.


Can you reproduce irreproducibility?

Give me a question which the LLM answers vastly differently on runs.

I keep hearing how it's dumb and wrong but no one ever shares the chat or prompt



Try this with ChatGPT or GROK or Claude

How many days of the week contain the letter d?

The answer I get with ChatGPT, and Grok is 3 and 6 with Claude.


I just used ChatGPT only, twice. Web interface in a Firefox private window, and in a Chrome incognito window. I asked them both the identical question "How many names of the days of the week contain the letter D?"

In Firefox I got 6. In Chrome I got 7. LLMs are not even self-consistent.

I have the screenshots if anyone cares.


What does this have to do with LLMs?

They stopped requiring SAT and ACTs in order to get a student population more representative of the population in general. This obviously allowed students that were not prepared for college into the system.

If you do well in your math SATs you'll likely do well in math college. SAT scores and college GPA are highly correlated. No idea why anyone thought it was good to ignore probably the strongest signal of success in college.


Isn't GP talking about PhDs here? If they're PhDs I imagine they've already had good enough GPAs/were smart enough to be selected for the program.

> Shareholders will have no say in how the company is run. If and when he makes a mess of things, they will find it very hard to sue him because of various waivers. The board – that entity that is supposed to look out for shareholders - is loaded with Musk loyalists.

Does anyone seriously believe that an independent board of investors can deliver better results than a founder?

If you look at the companies that built the most amount of wealth in the last 20 years, from Meta, Google, Tesla, Alphabet, Nvidia, what many of them share is more or less singular control by the people in charge. Sometimes its super-voting shares but other times its just founder mentality and ability to make big bets and set the direction.

The rest of the article is similarly non-sensical. Everyone will be forced to buy it but it's going to crash! The prior investors will sell their shares! The IPO is an exit mechanism!


Companies that are 80, 100, 200 years old or more have trouble with founders dying.

One of the disadvantages of relying on the founder is that founders die. If I'm trying to keep a fund going for the next 100 years, investing in a company that relies exclusively on a person who will be dead within 100 years seems problematic.


Alphabet owns Google, why did you list them seperately? Why is Microsoft not included? Why is Apple not included?

Pretty insane to then claim the article is non sensiscal.


Thats the definition of cherry picking and survivorship bias

I’d be genuinely curious about the data on this.

There are examples I can think of with a more traditional governance structure that did well: Apple, Amazon, Microsoft.


You're assuming cause and effect run in that direction, but arguably companies that do well are going to have stable leadership. Governance matters when things go of the rails. There is also the matter of how well these companies would have done with alternative leadership. They might be running well under capacity. Finally the stock prices are a terrible metric since they are based on sentiment. See various meme stocks.

Correct, everyone complains about "enshittification" but they fail to recognize that the most egregious cases occur when a company with a dedicated founder/leader transitions to rule by committee.

A lot of founders are the cause of enshittification. In fact, most of the big names I can think of are.

Marc Andreessen argued that we've already reached AGI. He says that the top AI models give better answers than 99% of people he has access to, and he has access to some of the best people in their field.

I'm getting more convinced. I mean, sure it makes dumb mistakes sometimes but its a particular set of self serving mistakes, commenting out tests in order to pass. We obv don't want this behavior but I wouldn't say it's dumb.

It'll be like the Turing test, which we just blew past years ago and no one cared. After all the hand-wringing about sentience and rights of the AI if it passes the Turing test, and now we just have AI bots running 24/7 writing slop.

How does everyone else feel?


> Marc Andreessen argued that we've already reached AGI. He says that the top AI models give better answers than 99% of people he has access to, and he has access to some of the best people in their field.

He stands to make billions if enough people believe him — unless you also do, consider that you’re the mark. For example, if that was true, it would have to mean that AI companies either aren’t letting customers use the good models or are instructing them to frequently make errors which reveal a fundamental lack of reasoning ability.

Consider also that his wealth means he hasn’t had to defend an idea stringently since the 90s. I wouldn’t be surprised if he does think LLMs give deep answers because it often looks that way until you critically review the response and ask questions like what’s missing which require you to have a decent understanding of the problem domain.


And you stand to lose your job and your identity as a programmer.

He makes billions but he already is a billionaire. Gaining billions more doesn't mean shit. The guy really has nothing to lose and the utility of what he gains contribute little to his life style.

I will tell you this. HN has been comically wrong about everything related to AI. They said driverless cars have no chance of becoming useable. Now Tesla FSD is almost there and I sleep in waymo cars. HN said AI will never code, now everyone uses it to code.

It's fucking stupid. This is one of the smartest forums on the internet but HN becomes next to stupid when predicting AI. Why? Because humans can't face the truth. When the victim of attack is yourself, it doesn't matter how smart you are... you have to scaffold a rationalization to spare yourself as the victim. You have to lie to yourself and tell yourself that you matter.

The truth of it is, while LLMs are not the end game, AI in general is on a trajectory to take over. It shows us how meaningless our skills are... not only as programmers but as artists. That beautiful song you felt had greater meaning? It's all reproducible via an algorithm because it never really had a greater meaning. It was just a pattern.


> Gaining billions more doesn't mean shit. The guy really has nothing to lose and the utility of what he gains contribute little to his life style.

You don’t become a billionaire because you aren’t committed to making a number go up far after you no longer have any significant unmet needs. He’s spending his life focused on business deals because that’s what he cares most about — if his true love was science, philanthropy, etc. he’d have been able to do that full time a couple decades ago.


Yeah an extra billion is nice. But they're not going through an crisis where there primary skill as a programmer is about to be annihilated.

Everyone on HN is in denial. A billionaire may or may not lie for business reasons. But their situation is nowhere near as critical as you or I.


Marc Andreessen has a strong financial incentive to feel this way and to convince others to feel this way.

I also think it’s easy to think that AI gives good answers if you don’t know the field well. In fields where I know the material, the answers are pretty variable and can be quite bad.


HNers have strong incentive to feel the opposite. Humanity in general has strong incentive to feel the opposite.

AI is not only replacing programmers, but art and the meaning of being human itself. It's showing us how trivial all of human creation is as it's just patterns from an algorithm.


>Marc Andreessen argued that we've already reached AGI. He says that the top AI models give better answers than 99% of people he has access to, and he has access to some of the best people in their field.

He has access to employees and yes-men. What he actually needs to hear, nobody will tell him, AI even less so. Every shit idea he has, would be "what a bright idea"-ed by both everyone around him and AI.

And of course there's the little matter that he makes money and increases his power by selling AI. What seller doesn't promote their stuff as the greatest ever?


> Marc Andreessen argued that we've already reached AGI. He says that the top AI models give better answers than 99% of people he has access to, and he has access to some of the best people in their field.

Investor with vested interest in AI companies makes claim of reaching "AGI".

He is one of the last people to listen to about AGI. Unless the term "AGI" means something entirely different to him vs to independent researchers vs to CEOs, since the term has become entirely meaningless.


Knowing the question is half of the answer. LLMs are great at scoping your context and answering precisely what you asked; it's also why they go off the rails when they misunderstand a part of your question. Incidentally, they're great at "knowing" and reaching for knowledge.

Humans have the advantage of perspective. We always lack some knowledge and answer broadly. This is bad if you have a particular goal in mind, but better if you're just generally learning, because you see more and learn to discriminate the correct from the wrong. And most importantly, being wrong is part of human ingenuity - because sometimes we turn something "obviously" wrong into something right.


He would tell you NFTs were AGIs if it might get you to buy them.

Getting the right answers is just half of it, you need to know the right questions to ask. I haven't yet seen AI crack that one.

[flagged]


I’m not an AI stan by any means and certainly no fan of Andreessen, but using the term “clanker” immediately biases your statement and can discredit what is a well-referenced or well-meaning comment.

Am I the only one thoroughly unimpressed by all these "scary" crafted doomer scenarios.

Yes, if you write a prompt that looks like a sci-fi book, the next token prediction engine will generate the rest of the script.

Can we all admit the AI will kill us crowd was wrong? I remember this crowd going nuts when we gave AI access to query the internet. They thought that was insane. At least we never gave AI access to control our computers... or run long running tasks... oh wait, all this stuff happened, and apart from these carefully crafted doomer scenarios, pretty much nothing happened.

1 billion users, many millions running AI non-stop in a while loop with full access to the internet, hardware and pretty much everything digital and ... nothing.


No, the agents are not being adversarially prompted here. Rather, it's a consistent failure across models of RLHF-based safety-pretraining not generalizing to OOD open-ended agentic computer-use settings, as I explain here: https://x.com/aran_nayebi/status/2061875809384538366

It's not over.

Qualcomm has come out and stated they intend to see networks of agents crawling live communications data in real time. It doesn't take a William Gibson to see where this can go.

https://www.theregister.com/personal-tech/2026/06/02/qualcom...


> Since then, traffic to DuckDuckGo has been booming. Last week, the company noted that web visits to its no-AI search page were up nearly 30% week-over-week, and its U.S. app installs were also up 18.1% week-over-week, with U.S. iOS app installs peaking at 69.9% week-over-week growth.

Of course there are no absolute numbers or scale. This is just an advertisement for DuckDuckGo. It's gross that previously respected tech publications run this kind of slop for clicks


DDG used to have a graph showing the number of queries they did daily. It was around 100 million searches a day before they removed it a few years back. They were receiving bad press at the time IIRC.

An educated guess is they're doing a similar number of searches today.


Oh, pity, seems they did kill that.

The traffic graph used to be at:

  https://duckduckgo.com/traffic
It was originally just a raw traffic graph, see <https://web.archive.org/web/20130711215117/https://duckduckg...>.

By late 2022 (18 November) it was a tabular presentation without the graph, and disappeared shortly afterward: <https://web.archive.org/web/20221118045948/https://duckduckg...>.

It's gone by 6 Dec 2022 (redirect to homepage): <https://web.archive.org/web/20221206080717/https://duckduckg...>


Isn't that the whole point of social media like Instagram? To convince the world you have disposable income and are living a lavish lifestyle?

Personally I like signaling that I have money. Why would you want people to think you're poor or cheap, except maybe when you're shopping for a car.


Why would you want to signal that? It relates in no way as to what kind of person you are. In fact, the richer you are, the more questions I have about how you got it and who got shafted along the way.

I wouldn't, you wouldn't, and countless millions also wouldn't.

But there is clearly a demographic who do use social media to signal lifestyle status, often using that lifestyle status to sell products of various kinds.

The erosion of enthusiast fandom into paid influencer "fandom" is whole subculture.


> In fact, the richer you are, the more questions I have about how you got it and who got shafted along the way.

What a sad way to live life. Not only because it's untrue (assuming you don't live in North Korea), but it's incredibly dark and destructive.


The fact that you think it's "sad" is a dead giveaway in terms of what you value in life and what you consider a virtue.

What do I value? I believe I live in a society where wealth is earned through hard work. It would be awful living in a society where I wrongly think everyone who has something I don't earned it through nefarious means, and there is no way to be an honest broker and create wealth.

This does not answer the question that was asked (and which I also wonder): why would you want to signal that you are rich?

For attracting members of the opposite sex and wanting to connect with other people of similar wealth as their values and interests more closely correspond with your own. I don't know if you noticed, but there are cultural differences across socio-economic classes.

Are you pretending not to understand?


They made a movie about people like you in the 90s, called "Clueless":

https://www.youtube.com/watch?v=ld_TueiQiQo&t=105s


> Why would you want people to think you're poor or cheap, except maybe when you're shopping for a car.

Personally, I don’t generally think about how other people perceive if I have money or not.


Ah, you must be one of those that cuts his own hair, shops strictly at thrift shops and wears bags as shoes.

I get my hair cut at a place where it costs $45 and I get a haircut every 1-2 months at best.

I wear my shoes until they break. I wear Vans Sk8 old skool high tops most of the time.

I am also on track to retire by 40 if all goes well. (Which is to say I don’t have THAT much money)


Ah, so you generally think about how other people perceive if you have money or not. Enough to answer an internet stranger inquiring about your wealth, and flexing about retiring at 40 (congrats).

Thanks for proving my point.


Maybe hthe GP is confident enough to have passed by the shallow indicator of displaying their wealth? The richest people I know personally (and who I deeply respect) absolutely delight in people under-estimating them, assuming they're "poor plebs" like the rest of us. They like money, recognize it's important and appreciate what it allows, but feel no need to advertise it.

I mean, people who say "why would I want status symbols" tend to not purchase designer clothes and $500 hair cuts. Because those are status symbols.

But bags as shoes is impractical. You want a sole to protect your feet.


Strongly agree. I would take it one step further, consider this line:

> Slopping out a 10,000 LOC untested Python/JS mess in 5 minutes helps nobody. The thought of this happening in every commercial environment simultaneously is horrifying.

You should def try to commercialize your slop and get feedback. No one cares about your tech stack or whether it's maintainable. Does it add value? You'll get a strong signal as to whether it does or not. And adding value, picking the right problem in the right domain, that's the hard part. You can always re-write or clean it up. Didn't Javascript start out as a weekend project? (maybe not a great example depending on how much you like JS)


The point of these broad ETFs is that they include everything. Let the market decide. Of course they should own one of the largest public companies in the world. They're changing the rules on inclusion because the ipo is unprecedented and not owning it because [reasons] would be a dereliction of their duties.

You want an ESG fund

Also I dont see how weapons companies are harmful. Unless you're so naive to think defense is not a thing any person or country has to worry about in 2026


I think their point is not the ESG component, but firms with traditionally irrational valuations (à la GameStop) for which index inclusion exceptions have been made to facilitate short term liquidity for IPO participants. Seems as though one should be able to hold the broad market less that component.

> Let the market decide... They're changing the rules on inclusion ...

You have the market deciding and the rules changing in the same paragraph and nothing's bothering you. I genuinely envy your peace of mind, my friend. Some of us are truly blessed.


The rules changed because it's unprecedented. It's not complicated. If your job is to "track the market" and there is this company that is worth $1+ trillion, you're not doing your job if you don't have exposure to this company.

Just be honest with yourself. The only reason you and others have an issue is because of politics. You don't like Musk for whatever reason and now you're very opinionated about the internal workings of index selection, when prior to reading about it in the NYT or something, you had no idea.

You don't care about the arcane byzantine process, you think rocket man is bad. I feel bad for people that get so easy manipulated. It's a hell of a way to live your life, waking up and reading corporate media to tell you what you should be angry about today


Ignoring the moral questions, the issue is that the old waiting period had a purpose. The market *isn't* getting an opportunity to decide here. There's a concern that these are a glorified pump & dump where initial investors extract all the value via the IPO and then the price craters once it's on the open market.

It'd be nice to give the market a period of time to figure out what it *really* values these companies.


> moral questions

Do you hear yourself? This is about index inclusion.


I stated that because most of this thread is on that topic and specifically didn't want it involved for this. I'm sorry if you're too biased on the topic to take that into account.

Doesn’t matter that it’s unprecedented by value, Whole point of rules is confirm that value isn’t fake by letting the market stabilize after IPO and then if value is there, it’s added to the index. Yes, some money could be lost if value is there but reverse is true as well.

Hang on, wasn't the market supposed to work it out? How come it's someone's job to fast-track inclusion? Isn't intervening directly in the market a communist policy? Are you a communist by any chance?

> Let the market decide. Of course they should own one of the largest public companies in the world.

The pension fund is the customer here. The market is already deciding. You're free to invest your own money as you see fit. The pension fund's money is not yours to decide what to do with.


Let the market decide what, though? What the market cares about may be different from what you care about, if the average investor has a higher tolerance for risk that you do. For pension funds, long term stability is key. A wide spread of large companies has traditionally been a good way to achieve that, but that isn’t guaranteed.

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

Search:

HN For You