Setting aside how shortsighted it is to fire your employees to replace them with AI, Ford also screwed up by firing the wrong employees. LLMs work best in the hands of experienced senior engineers who can work at a high level of abstraction because they already understand all the pieces underneath.
In a sense, using an LLM agent is like providing instructions to a very smart, very quick junior who despite being brilliant has some blind spots and lacks institutional knowledge. That's something that seniors excel at, so by firing your seniors you've fired the people best positioned to make full use of LLMs.
That's just the basics. To craft a prompt for a complex architectural task, you need to know the solution at least on an abstraction level. If you don't have the right system design in your head, no llm is gonna conjure it out of thin air
> In a sense, using an LLM agent is like providing instructions to a very smart, very quick junior who despite being brilliant has some blind spots and lacks institutional knowledge. That's something that seniors excel at, so by firing your seniors you've fired the people best positioned to make full use of LLMs.
aye.
I've posted this here at HN several times but I had my intern try to track down how many CVEs from a list of vulns we found were being exploited in the wild -- couple years ago, pre mythos that is. I also took the list to Copilot and Claude.
All 3 got different answers, albeit off by one or two. The intern told me at least he didn't know about X, which was far more useful. I later had him whip up a plan and some basic code to patch some of them, and the experience comparing his answer to Copilot was similar to before as well -- both mostly worked, but didn't, and in different ways, and mostly due to not knowing institutional best practices.
"Ford rehires 350 engineers after AI fails to preserve expertise or train juniors" In order to rehire someone they must laid off or fired? You don't rehire new employees?
I think most of them were losses by attrition. Where they don't replace lost employees. That's usually the preferred method of downsizing if you can get away with it.
I wonder if Nico will be feeling so cocky when Papermark gets their general counsel involved. The public Twitter shaming was clearly an attempt to resolve this without litigation, but hey, if that's how Nico truly feels, guess he gets to see what's behind door #2 (a massive bill for a legal retainer).
> it's pretty clear that YC does not care about a negative reputation.
Perhaps not what the general public thinks, but I assume YC cares a lot about its reputation among VC firms that fund its companies, because VCs don't like being scammed (directly, or indirectly through unknowingly funding scams)
Many YC companies do bad things, and I guess they do so independently. There may well be repercussions for the most egregious cases, but I suspect a lot of ill-behaviour simply flies under the radar.
For example only yesterday I got spam from an YC company, Polymath, and I replied back asking where they got my details from - no response yet. Once I get something I'll make a GDPR subject access request, then a deletion request. I hope the overhead of that causes them to rethink their spamming campaign.
My comment was not about doing a generic bad thing - it was about scammy behavior in particular (which ties to the Delve incident). YC depends on the VC ecosystem to fund its companies, and no VC wants to be scammed. If a reputation of cultivating/condoning/obliviousness scammers takes root, that would be bad for business.
> But I'm not going to complain to YC about it.
I am not complaining, or even expecting a moral decision. I'm legitimately curious how this will shake out, for purely capitalistic, reputation-management reasons.
I have also gotten spammed by a YC startup, but they spammed an email that I use in git commits, and lead with "I saw your fork of $POPULAR_PROJECT, pretty cool!" or something like that and then continued to pester me with their drip program even as I replied asking them to never email me again.
Good luck with referring to GDPR. Try clicking through YC startup list and see how many load GA and other trackers onto their landing pages without a consent banner or even a privacy policy sometimes. It’s baffling.
I didn't realise that one could forcibly require a competitor to disclose trade secrets.
Now, INAL of course, but I would think this sort of mechanism would be quite gameable from both sides ( i) a wealthy competitor legally forcing a promising upstart to reveal source ii) a copycat working out some kind of arrangement where the code itself is licensed to them via shell company based overseas.)
As with most legal hacks, the courts figured this one out long ago :).
If someone is trying to dig into their competitor's trade secrets via discovery, the court offers multiple ways to safeguard against that. The defendant can identify information as a trade secret and ask that it be protected in some way - for example, the documents may be restricted to "Attorneys' Eyes Only", so while the plaintiff's attorneys can review the material, the plaintiffs themselves are barred from reviewing it. Or the judge themselves may get involved in an in-camera session.
There are software engineers that specialise in source code analysis that lawyers will often use in these cases. The engineers will be given access to source code in secure environments where they're not allowed to bring any device in or out. They review, analyse, and write up a report using pen and paper, that can then be reviewed by the lawyers.
Absolutely. It was very similar to one of my first jobs: "Legal Technical Analyst". Not as much time doing deep source analysis, but basically translating things for lawyers: "So as far as this claim of copyright/plagiarism... this block here, that's CS 101 stuff, that block there, that's novel, and does x, y and z".
The tweet was fine - it was directly addressing Corgi's claim that they had "vibe coded" DataRoom when they had copied and pasted it from Papermark. The problem is the OP chose to perform a contextectomy on the tweet and make it look like it's making a completely different argument.
Yeah, the title that the OP chose is so sufficiently misleading that I think this one will need to be get changed by the mods. Seitz isn't opining on the ethics of vibe coding in his tweet, he's pointing out that Corgi literally just stole Papermark's AGPL codebase and passed it off as vibe coding.
Short segments of popular works sure. Many UI pages with identical layouts and copy, essentially zero chance. The agent had access to the original code at inference time.
It's nearly word-for-word the content of the tweet. Right at the top. It isn't misleading unless you literally don't even bother to open the linked content.
Just ban users who comment without reading, I think that would go further to keep the quality of discussion high.
The number of bots/trolls responding to the title without reading the content and missing the point entirely is astounding, honestly, and I don't think any of those posts are contributing to high quality discussion. We could do without those users.
"but but but I can't/won't open twitter links" - then don't flap your yak-hole. Ignoring for a moment that the content has been reproduced in full in this thread, and another user has provided an alternative xcancel link.
Ideally yes, but we know people don't RTFA - there's a reason that initialism dates back to early Slashdot.
The paraphrase is doing a lot of heavy lifting to convert it to ragebait. Had the OP gone with something like "you didn't vibe code it, you plagiarized Papermark's open source project" (may need some editing to fit under the character limit) it would have at least been more true to the original tweet.
I know I RTFA, and I know I'm not interested in discussing things with people who don't. Maybe others feel differently, because more people is better or something. Information pollution is a serious, persistent, growing problem and I'm just not inclined to be tolerant about it anymore. Mistakes are one thing, deliberate stupidity is another.
If you come to book club without reading the book, and you derail the conversation into something completely irrelevant, you're not getting invited back.
Canadian firms can easily access U.S. capital markets. So the question remains of why we aren’t building all kinds of data centers out in the tundra next to giant hydro plants.
You get access to a whole bunch of bleeding edge open models including GLM-5.2, Kimi K2.7, DeepSeek 4 Pro, etc. Inference is run on US/SG/EU cloud providers with zero data retention policies. The $20/mo tier is very generous, in my experience.
Well I tried the $20/mo tier and used GLM specifically and did maybe 3-4 hours of work and I'm already through 50% of my monthly tier and blew through my time limited quota twice. I won't renew for another month.
Which I think only underscores my point that actually the GLM models are not very cost effective.
They essentially cost the same as the SOTA models from OpenAI and Anthropic, while not being quite as smart. I could have gotten about the same amount of work done on the $20 Codex plan. And I had to use my $100 Codex plan to finish the work GLM started before it ran out of quota. And also to fix it since GLM left a bit of a mess.
I like that GLM exists. Other Chinese models are far more cost effective. GLM is expensive, even on a fixed plan.
Ollama can’t meaningfully subsidize their subscriptions - there is no business case to do so because they are a commodity host. If you want to compare subsidized subscription value you would need to compare with z.ai’s plans. One problem with any comparison is that they are all very opaque in terms of usage and the plans change a lot over time. I got on pro at $30 a month so it’s a very good value - compared to $20 Claude/Codex plans I get at least 10x the usage and I use all 3 regularly. At today’s prices Codex pro ($100) is likely a better value.
But if you are building a product or in an enterprise environment where you essentially have to pay API rates then GLM is the best value hands down.
There's a blanket statement at the bottom of the pricing page, which I would hope also applies to GLM-5.2:
> Where are models hosted?
> Ollama hosts models and compute resources primarily in the United States. To serve global demand, we may route to Europe and Singapore for additional capacity.
> Is my prompt or response data trained on?
> Prompt or response data is never logged or trained on.
> Who does Ollama partner with to host models?
> Ollama collaborates with NVIDIA Cloud Providers (NCPs) to host open models.
> When Ollama partners with providers, we require no logging, no training, and zero data retention policies in place.
> It's always a shock to me how opaque most other models are!
This is (unfortunately) by design. The proprietary models hide their reasoning traces so they can't be used for model distillation. Sometimes even when they do show reasoning, it isn't the model's real trace - IIRC, someone was able to demonstrate that Opus' reasoning is usually a summary made with Haiku behind the scenes.
It is such a momentum killer being forced to stare at a silly word for 4 minutes instead of being able to read the thinking as it streams in. I can’t wait until I can drop Anthropic at work. Their UX sucks, intentionally, for anti competitive reasons like “don’t distill our model we trained on all the data & IP we stole and processed with the mass exploitation of data workers in the global south!”.
Agreed that you can export-control the closed source models pretty well (although I think the administration is gravely underestimating the long-term damage that will do to the US economy).
The bigger problem will be if someone (such as a Chinese AI lab) releases a Fable/Mythos-class open weight model. That you can't really export control successfully. Sure, you could class it under EAR or ITAR, but that's just going to make using it difficult for American companies, not everyone else. It would be a stupid protectionist measure that would only hurt the US - so I fully anticipate the admin would try it.
I can't think of anything better that could happen than a Mythos/Fable level model released with open weights. It would be a huge step forward for the whole world.
Particularly if it came paired with DeepSeek pricing.
Alternatively, we could try to keep closing barn doors after the horses are already in the next county, have been hit by a car, and are on their way to the rendering plant.
None of your statements follow eachother. Anthropic did a few premature optimizations so they could immediately show their ineptitude. If it wasn't their constant advertising (safety focused lab whose model escapes their sandbox every other week with an unlisted exploit, seem to leak models or proprietary code every other week). It would be their insane pricing for a basic utility, intentionally locking the rest of the economy (And most major corporations) out of the (minimal, cos it's AI not electricity) gains. Of course there was never actually a mote. It's a freaking machine learning model. GLM5.2 was trained on a tenth the hardware Mythos was trained on. If anything Anthropic are already 6 months BEHIND. They are BEHIND google in terms of true multimodal intelligence as well. They are ahead of their partner in crime OpenAI.
Opus and Codex are both catching plenty of very good improvements to my GLM plans. It gets a lot right too, has a lot of good things it does, good habits and practices. But it's not as smart, not as observant, not as able to craft a nice system. In my experience.
It'll depend on what law they're restricting it under. The obvious play would be to put it on the Commerce Control List so it's covered by the EAR (Export Administration Regulations). If so, compliance is pretty well-understood, just a giant pain in the ass that'll pretty much limit use of these models to companies that already have EAR/ITAR compliance offices.
And Canada's hate speech laws are pretty reserved compared to some of their European counterparts. You pretty much have to be an actual goose-stepping Nazi or white robed Klansman to catch a charge.
In a sense, using an LLM agent is like providing instructions to a very smart, very quick junior who despite being brilliant has some blind spots and lacks institutional knowledge. That's something that seniors excel at, so by firing your seniors you've fired the people best positioned to make full use of LLMs.
reply