mtrovo

I have the totally opposite sentiment from the previous LMGTFY meme. It was easier before to ask humans rather than do some research using Google, that's not true anymore with LLMs.

Now if I ask a question to humans I'm looking for their personal opinions drawn from their personal experience, be it ticket size for a feature on a legacy codebase, a design decision or a product strategy decision.

People can host these models is doing a lot of lifting here, these are models that depends on 5 digits on specialized installation to run on.

IMHO, this has the impact of softening the impact of data centers sitting unused in the long term if they can still serve open weight models, even if Anthropic or OAI have to scale down their expansion rate to pay the bills.

Regardless, reality has to give at some point; these valuations don't make any sense. We've been valuing GenAI as disruptive work, when in reality they're much closer to cloud providers with a beefy, one-pony-trick R&D department.

It's a strange realization that this looks so much like LinkedIn because it actually is LinkedIn.

It seems that an LLM is deliberately being forced to generate in the style of LLM slop.

Considering their current valuation and the prospects of getting any of this money back, that's a genius exit strategy.

Yes

That might be true right now, but how long until you have to move the goalposts? In my experience with DeepSeek and Kimi, they're as capable as the frontier was four months ago, which already solves a big chunk of the coding tasks that I'm interested in.

Right, but I'm pretty sure this is not the first time somebody thought about solving it.

https://xkcd.com/927/

He's the "head of strategic futures" of the 1T valuation company based on fear and vibes, I think he's doing a very good job at it.

20% efficiency means you can open single isle flights on 25% longer distances, which open a lot of routes that are not profitable with regular long haul planes.

I've flown on the A321LR a few times, and it's always crazy to think about that tiny plane cruising over the Atlantic without stops.

That could pay for a lot of tokens.

Which on itself is wild. The main issue with the discussion isn't a rift in capabilities, but a rift in the valuation model itself.

If China can match the pace and produce frontier models at a fraction of the cost of US alternatives, even without access to SOTA hardware, what is the competitive advantage for the current valuations of Anthropic and OpenAI?

You're half joking but this actually happens already. As you can imagine there's a lot of backlash on dumping good clothes on Europe itself so they export them in bad conditions just to have it burned out of sight.

https://changingmarkets.org/report/trashion-the-stealth-expo...

And it's not just old clothes being discarded, another related study showed that around 30% of clothes returned from online stores are not even looked over to see if they're worth selling again and are discarded straight away.

That's a really cool concept, this could be awesome as a tutor with diagrams and stepped animations.

I have a remarkable and love it, but I think there's still so much potential for new forms of interaction on eink devices. The lack of dev kits and their price just makes it a big niche for now.

> A good argument is that therefore there is no reason to pay royalties on AI generated work as it is the equivalent to public domain.

That has some different second order consequences that I don't think you're seeing. It's not that they will be free to you as a user, it's more that they will be free from the platform perspective to do whatever they want with the revenue they get from it.

Say for example you have a platform with Spotify monetization scheme for instance, which is already very unfair to small artists. But now imagine you have to compete to be included on auto play or playlists against something that's basically free for Spotify, what's your chance of getting any money out of it? Say Spotify changes their algorithm and starts pushing 20% of all auto play playlists to consist of AI songs, that's basically a 20% bump on their profit basis.

We're talking about a 400k dollars car, maybe they could find a way to add this expense into the design.

Youtube Music is quite good for what you're describing.

All the shots at the name apart I think this is a very good strategic move. The other frontier labs would die to have this level of surface available for their models as a testing ground, with the current state of things on Apple side the ChatGPT on MacOS integration is probably the best everyone will get for a good time on how a full integration of LLM model with OS could really looks like.

Agents will need a different level of understanding of your activities across different surfaces to act effectively, IMHO the OS is the perfect place to offer it.

> and limited to the Google Play Store too, which just isn't good at this point.

Care to elaborate? I have no ide a what you're talking about here.

Most of the issues are around code hygiene rather than just LLM code being bad. You're creating code 10x faster, but you're also writing unit tests 10x faster, not just that but integration tests, CICD workflows, prod monitoring, product and engineering documentation, etc. It was already the way to get good code quality before, nowadays I think it's just reckless to generate code that's not backed by 100% test coverage and pass all lints and static checks configured.

It's the 3rd or 4th of threads like this in the front page and it's still not clear to me what are the alternatives that privacy advocates vouch for? Dead internet theory is happening, you have botnets with more budget than most of the third world countries and you could also add openclaw usage to same bucket. There's a real need for a protocol or specification for how to attest that an action was really done by a human and that human can be proven to be the one the service provider think they are. I don't think cryptography by itself would solve that right now.

LLMs fail at laser-focused troubleshooting, but they excel at brute-force breadth. Priming an agent to list 50 distinct possible causes for a database connection failure and investigate each one of them works better than hoping it guesses the exact root cause.

The main take away from this article for me is that battle scars can be used to unbog these agents. That explains the current productivity divide we're seeing, seniors use their past experience to unbog agents. Juniors naturally frame the problem into the brute-force breadth approach. The problem then is focused on Mid-career devs that get the worst results because they are naturals on framing the problem on brute force way so they try to force agents through rigid logic without deep experience.

The interesting thing about religions as a whole is that the timespan is so big that you can really see how the backbone of the narrative stays the same while the fanbase and how they pick winners changes a lot, the Vatican state itself is a theocratic state created by an agreement between the pope and the Mussolini.

And if you wanna go back even further just remember that while Europe and christian countries were living in the dark ages the Islamic world was the one driving forward scientific knowledge and the exchange of ideas with the East. https://en.wikipedia.org/wiki/Islamic_Golden_Age

Behold VikTok, the new competitor to Moltbook, soon to be acquihired by Oracle.

I think the main issue is treating LLM as a unrestrained black box, there's a reason nobody outside tech trust so blindly on LLMs.

The only way to make LLMs useful for now is to restrain their hallucinations as much as possible with evals, and these evals need to be very clear about what are the goal you're optimizing for.

See karpathy's work on the autoresearch agent and how it carry experiments, it might be useful for what you're doing.

Clothes, wristwatches, cars, you name it. It's a very common play on luxury brands, Hermes Birkins is the most famous that comes to my mind and follow a very similar playbook.

Apart from the KYC aspect of the process it's their way of solving the problem of artificial scarcity on the second-hand market as the article explains. They want a second hand market to exist to indicate that this is a luxury item, but too many and the price tanking with excess supply.

> This document was written by an LLM (Claude) and then iteratively de-LLMed by that same LLM under instruction from a human, in a conversation that went roughly like this

This is hilarious.

The hardware looks fine, but Apple's software vision is so confusing.

MacBook Neo is cheaper and weaker than a MacBook Air, yet shares the same price and single-app mindset as an iPad. It uses a phone chip similar to an iPad Pro, but gets multi-user support and a keyboard.

I struggle to run Tahoe on my 16GB M2 Air and somewhat I have to believe running it on a 8GB phone chip is gonna be alright, which if true have me thinking what exactly is the role of iPadOS anyway.

Ultimately, it feels like iPadOS and Tahoe are on a crash course for a middle ground that nobody asked for.

At least that's the story LLM labs leaders wanna tell everyone, just happen to be a very good story if you wanna hype your valuation before investment rounds.

Working with LLM on a daily basis I would say that's not happening, not as they're trying to sell it. You can get rid of a 5 vendor headcount that execute a manual process that should have been automated 10 years ago, you're not automating the processes involving high paying people with a 1% error chance where an error could cost you +10M in fines or jail time.

When I see Amodei or Sam flying on a vibe coded airplane is the day I believe what they're talking about.

> Workforce re-skilling programs should prioritize “fusion skills”, such as prompt engineering, data stewardship and human-in-the-loop decision-making that enhance human-AI complementarity.

First time reading the term fusion skills used in this context and I really like it.

I created something very similar but to display raindrop links into the remarkable and sync highlights back into raindrop. I also added a GenAI powered paper summary filter as a preface to the papers it send to the remarkable which is working quite well.

As you mentioned Remarkable file format is a PITA to extract highlights, one thing that helped a lot was to add an OCR fix phase that uses Gemini flash model to fix common OCR errors and to merge single highlights that are across pages.