> Some emerging research suggests Earth’s magnetism matters, as well. Being barefoot on the ground versus 30 stories high in a downtown skyscraper puts you in a completely different magnetic environment.
Really? My instinctive reaction to this statement is that it's quackery. Would be very interested to see headlines appear about this if it's true.
It's definitely quackery. There's way more "natural" magnetic variation between two different cities than between a ground floor apartment and a 30th floor apartment (see magnetic declination maps), and that's even before you get into the way-more-powerful magnetic fields generated or present in a home.
This is called "grounding" or "earthing" and Wikipedia doesn't pull punches in dismissing it [0]. You can buy grounding wires for your mattress or copper rivets for your shoes.
From my tissue engineering class in undergrad my 2 cents is this is maybe 95% supplements quackery and 5% genuine mystery. You can find papers that claim EM fields have cellular interactions [1]. I don't think serious scientists will touch conventional earthing, though.
People already live in different altitudes, ie with vastly different amounts of rock between themselves and the bits of the earth that produce her magnetic field.
So colour me skeptical of this claim as well.
And while I like going barefoot for other reasons, I don't think it makes much of a difference to the magnetic fields.
> People already live in different altitudes, ie with vastly different amounts of rock between themselves and the bits of the earth that produce her magnetic field.
It's not really the same thing. Construction and especially the high content of charged metal (whether it's power conduits or plain old frame) may make a considerable difference.
Do many people sleep in such tall buildings? Obviously they exist, but I can't think of any over-20 story residential building I've ever been in. Office space, sure, but not residential.
I would not jump to such conclusions, I would assume is not fully known and needs to be further understood.
on the other side, do you know we have polarized and non-polarized cells? do you know they can be affected by electromagnetism?
this two things already establish the existence of a relationship and effects between them. If we then consider that cells mechanism work based on differential of electrical or chemical potentials, and that they continuously use that for their functions, reactions or environmental compensations well...
I just think they are related and we need to understand better.
Fascinating, but polarization is a red herring. Magnetism is famously the wildly-unreactive half of electromagnetism; copper is REALLY sensitive to electricity and REALLY insensitive to magnetism.
The methane will eventually break down and be part of new plants, but in the meantime it will heat the planet much more than if it hadn't been there. Think of it this way: industrial scale cattle farming essentially makes and feeds biochemical reactors that turn hydrocarbons into a gaseous form that maximizes their global heating effect. Consider the counterfactual where we don't stuff billions of fart machines full of corn.
> The methane will eventually break down and be part of new plants
To be exact, excess methane go in the high troposphere, where it is broken down in CO2+H2O (the half life is very short, I don't remember which is which but it's either 5 or 20 years). So in any case, it accelerate climate change.
Are you implying that's a bad thing? Doing things for social status has been an integral part of society for millennia. It's a legitimate motivation that can benefit both the recipient of the status and the rest of society. In this specific case, if you assume that the author makes content that is useful to you only because of the social status reward, taking away that reward means the author will stop making content that is useful to you.
What do you get out of that? The utility of having the end product (a generated game for an obsolete console) seems pretty small, and personally I would think the fun is in learning about the inner workings of the time capsule that is 90s hardware. Would you mind explaining?
It's just not the same. I made games for the PSP and Vita and it's "different" than making games for Windows or Playstation/Xbox. There's something about seeing your creation on a tiny handheld. I guess maybe nowadays we got desensitized to it because our phones are actual supercomputers which can run Fortnite without breaking a sweat, but making a device with 2MB of ram play your game as you imagined it is just hyper cool and (for me personally) is incredibly satisfying.
I'm not trying to be dismissive here - I am the type to be into the nitty gritty nerdy stuff that using Claude skips, BUT I have exactly zero experience developing anything for a legacy hardware platform in any capacity (with or without Claude) so genuinely curious to learn varying perspectives here.
Is it about the idea of the platform having limited capabilities (e.g. akin to Game Bub or MiSTer) or about tactile aesthetics (e.g. possibly vaguely comparable - though not really - to r36t or play.date) or something different entirely.
Honestly I think part of it is just that it's cool to see software running on this platform. I remember coding some trash apps on the 3DS back in the day simply because I could. It was great seeing code running somewhere it isn't strictly "supposed to"
If you enjoy this sort of thing I’d highly recommend pico-8 and tic-80. Both have pretty full featured dev environments and have a set of restrictions that make developing for them fun.
If your game fits in the bounds of N64 hardware it makes perfect sense IMO. You basically get virtually all modern platforms for free. Much easier than wrestling with all the modern API that are "platform independent" but not really.
While I agree with you, I think it's interesting that LLMs have progressed enough to be able to write working code even for niche and heavily-constrained platforms like the N64. I'm sure it helps that modern toolchains make that easier (I don't expect very many LLMs to be able to put up with the old Windows 9x or IRIX toolchains, to say the least!).
I mean, there is plenty of people out there who enjoy the process of making a game(they have a really clear idea of gameplay mechanics they think would be fun when put together) but they really don't enjoy programming at all. I guess it's kinda the same - I Want to see my game happen, I don't want to code it. Which isn't that different from where all the other tools are leading us anyway. Nowadays you can make a video game by just downloading an engine and building everything using the included scripting language most of the time - no need to understand rendering, IO, input or anything else.
> Nowadays you can make a video game by just downloading an engine and building everything using the included scripting language most of the time - no need to understand rendering, IO, input or anything else.
I don't disagree, but I don't see it as a positive, looking at the current game releases and their performance being shockingly unoptimised, there is a lot of lost real world value in knowing the lower level details, it's like anything really, sometimes you have to do the hard/unappealing work to create your best work.
Hmmm look, I can offer you my perspective on this.
I've been a low level engine programmer in video games all my working life. I worked on a few AAA games that sold 20M+ copies, I'm not saying this to boast, I'm saying this to say I know how the sausage is made. My literal day to day job is fixing issues like "The RHI thread on the Switch takes 0.5ms longer than it should" and optimizing things as much as possible.
I think if you asked me few years ago, I would have also taken the same stance you did - that these tools make it "too easy" and we get unoptimized crap out there.
But you know what, nowadays I feel like I mellowed out a lot. A lot of these so called "friendslop" games are horrible in terms of technical work. And you know what? They still bring joy to people. They still make people laugh and have a great time with the people they like. One of my favourite memories from a game I worked on was reading comments from people who said they were looking forward to just playing the game after work in the evening with their friends. Was optimizing the IO performance on PS4 essential to make the game happen? Sure. But what mattered more was that the game was actually fun to play.
Nowadays I see the improvement in tooling as nothing but a positive. There will always be hardcore engine programmers who know how to do this stuff, I have zero doubt about it. But allowing people to just download an engine and make the thing is absolutely fantastic. To say that it's bad - to me personally - that's gatekeeping. Saying you can't make a game unless you know how to write a renderer or a physics system is so offputting to people who want to make a game but just don't know how - we should be encouraging them, not chastising them for it. Little Big Planet did it well as one of the first games of that kind, if you recall that - just saying, here's a sandbox, make a thing that brings joy to other people. Isn't that what gaming is about?
And on a more serious, business side - it's kinda crazy that Epic lets you use UE for FREE until you start making serious money on the game that you made. You can start a company, hire people, and just start making the game you want to make without paying anything for the tooling. Very few other companies in the world let you do this.
You know thats completely fair. I'm not in the game industry but I see some friends and what they do and I do have to admit, I wouldn't be able for it so it's not really fair for me to critisise.
> it's kinda crazy that Epic lets you use UE for FREE until you start making serious money on the game that you made.
Thinking about how powerfull UE is, it is kind of insane.
One I've started noticing is the "quotation marks around a phrase awkwardly trying to bundle a concept" thing. Like all LLM cliches it's something that has been used in writing for a long time, but I've seen it so much more recently. I think lots of people have picked this up from seeing LLMs use it. But like you said, who knows.
> It's literally the free market at work--it's really simple, if you don't believe in whatever he says, don't speculate!
This, too, is actually addressed in the article - the whole point is that you're being forced to participate/speculate through this rapid inclusion in indices and mutual funds, because he's had the indices change the rules for him. So no, it's not that simple.
The indexes/mutual funds have a 1 year limit. The company would've been introduced into them anyway. You think SpaceX is going to collapse within a year? They fast forwarded the inevitable. But I do conceded that is shady to do. But to me, that seems to be the _actual_ main complaint of the article. Everything else is "i don't really like Elon Musk and don't believe in his vision"
Same, but recently I discussed the LLM flood on Reddit with some friends IRL and they were surprised I had left Reddit, as they simply hadn't noticed significant bot activity. Mind you, these are very AI-aware people. I have little hope that normies will catch on and actually mind enough to leave.
Something I’ve noticed is a lot of Gen Z people just straight up giving up phones and social media lately. Not so much because of AI but just the widespread negative impacts of what the tech has become.
There is a danger in chronic abuse resulting in upregulation. Mixing the two at once is no problem for the liver, which is also why patient information leaflets for paracetamol do not contain a warning to avoid alcohol, only about chronic alcohol abuse.
Your crappy source is vague in what consumption pattern constitutes a risk and actually cites a better source that supports the idea that acute alcohol consumption reduces paracetamol toxicity. https://www.biorxiv.org/content/10.1101/2020.07.07.191916v1....
That's a mathematical model, but this relationship between the two is what I was taught in medical school and it is still supported by the science. There's plenty of other sources, I just picked that one because your article cites it. Just search for "paracetamol ethanol" on Google Scholar.
These are good practices to keep in mind when setting up GenAI solutions, but I'm not convinced that this part of the job will allow "data scientist" as a profession to thrive. Here's my pessimistic take.
Data scientists were appreciated largely because of their ability to create models that unlock business value. Model creation was a dark magic that you needed strong mathematical skills to perform - or at least that's the image, even if in reality you just slap XGBoost on a problem and call it a day. Data scientists were enablers and value creators.
With GenAI, value creation is apparently done by the LLM provider and whoever in your company calls the API, which could really be any engineering team. Coaxing the right behavior out of the LLM is a bit of black magic in itself, but it's not something that requires deep mathematical knowledge. Knowing how gradients are calculated in a decoder-only transformer doesn't really help you make the LLM follow instructions. In fact, all your business stakeholders are constantly prompting chatbots themselves, so even if you provide some expertise here they will just see you as someone doing the same thing they do when they summarize an email.
So that leaves the part the OP discusses: evaluation and monitoring. These are not sexy tasks and from the point of view of business stakeholders they are not the primary value add. In fact, they are barriers that get in the way of taking the POC someone slapped together in Copilot (it works!) and putting that solution in production. It's not even strictly necessary if you just want to move fast and break things. Appreciation for this kind of work is most present in large risk-averse companies, but even there it can be tricky to convince management that this is a job that needs to be done by a highly paid statistician with a graduate degree.
What's the way forward? Convince management that people with the job title "data scientist" should be allowed to gatekeep building LLM solutions? Maybe I'm overestimating how good the average AI-aware software engineer is at this stuff, but I don't see the professional moat.
I don't really see why evals are assumed to be exclusively in the domain of data scientists. In my experience SWEs-turned-AI Engineers are much better suited to building agents. Some struggle more than others, but "evals as automated tests" is, imo, so obvious a mental model, and can be so well adapted to by good SWEs, that data scientists have no real role on many "agent" projects.
I'm not saying this is good or bad, just that it's what I'm observing in practice.
For context, I'm a SWE-turned-AI Engineer, so I may be biased :)
I think there's a lot of methodological expertise that goes into collecting good eval data. For example, in many cases you need human labelers with the right expertise, well designed tasks, well defined constructs, and you need to hit interrater agreement targets and troubleshoot when you don't. Good label data is a prerequisite to the stuff that can probably be automated by the AI agent (improving the system to optimize a metric measured against ground truth labels). Data scientists and research scientists are more likely to have this skillset. And it takes time to pick up and learn the nuances.
I agree with you take the there isn’t a lot of specialist work for data scientists to do with using off-the-shelf LLMs that can’t be done by an engineer. As an AI-aware software engineer myself… this stuff wasn’t that hard to pick up. Even a lot of the work on the Evals side (creating an LLM judge etc.) isn’t that hard and doesn’t require serious ML or stats.
But aren’t there still plenty of opportunities for building ML models beyond LLMs, albeit a bit less sexy now? It’s not like you can run a business process like (say) AirBnB’s search rankings or Uber’s driver marching algorithms on an LLM; you need to build a custom model for that. Or am I missing something here? Or is that point that those opportunities are still there, but the pond has shrunk because so much new work is now LLM-related? I buy that.
> I agree with you take the there isn’t a lot of specialist work for data scientists to do with using off-the-shelf LLMs that can’t be done by an engineer.
Conversely, data scientists are doing software engineering, including webdev. It’s an interesting time. I think it’s less about the job title demarcation now, and more about output.
I agree. It is difficult to convince leadership to do this work at all ("it works on my example, ship it"), and in my experience most DS don't even want to do it.
One of the key value is that it forces some thinking about what is the task you want to solve in the first place. In many cases, it is difficult if not impossible to do it, which implies the underlying product should not be built at all. But nobody wants to hear that.
Doing eval only makes sense if making the product better impacts something the business cares about, which is very difficult to do in practice.
I think most use-cases will still use simpler models like XGBoost etc. rather than LLM's. Customer segmentation is a really common use-case with no need for an LLM. Same for revenue/LTV forecasting.
Perhaps they can use the LLM to write and deploy these models without needing a Data Scientist but that seems risky to say the least.
In my company, the most Data Scientist-adjacent people are the Data Analysts but they tend not to have programming experience beyond SQL and basic Python and they aren't used to using the terminal etc.
Do those use cases need LLMs? Probably not. but if good results can be had with a day of prompting (in addition to the stuff mentioned in the article, which you have to do anyway) and a smaller model like Haiku gives good results why would you build a classifer before you have literally millions of customers?
The LLM solution will be much more flexible because prompts can change more easily than training data and input tokens are cheap.
I don't disagree that very numerical tasks like revenue forecasting are not a good fit for LLMs. But neither did a lot of data scientist concerns themselves with such things (compared to business analysts and the like). Software to achieve this has been commoditized.
I don’t actually even know what people are hinting at when they say that LLMs replace the need for building custom models. Regression models? People are using LLMs instead of say building a Bayesian hierarchical model? That’s not possible. Time series modeling using an LLM? Also ridiculous. Recommender systems? Ok maybe, still utterly ridiculous and abysmally slow.
For anything NLP sure, it definitely wins. However, I’ve just recently used some big fancy OpenAI model to actually just label thousands of text data for me, just so I could build a classifier with CatBoost. Guess what, inference speed is at a guaranteed sub 100ms and it costs $0 in tokens. The”AI Engineer” solution here would be just run every classification request through an LLM.
AI Engineering is going to have the same problem we had when Data Science as a term arrived and you had every Statistician saying they’re just re-inventing everything that exists in statistics, poorly.
You're right. For years the real impediment to "AI" products at many companies was the sheer crappiness of ML frameworks which were built by and for grad students, not professional engineers.
When LLMs appeared it was just so much easier to use then as an uber model and leave behind the training and inference infrastructure (if you can even call it that).
Now that LLMs can code I expect we'll be coding up custom model pipelines more and more... but only when we stop subsidizing LLMs.
One thing data scientists brought to the table was statistical rigor in the models, but that seems to have left the building at this point with LLM-based solutions.
As a AI-aware software engineer currently creating systems that integrate with LLM provider APIs for my company- who also has no idea what an eval is or how a data scientist thinks about RAG. I honestly don't see what value a data scientist would bring to the table for my team. Maybe someone would care to enlighten me?
You recognize that you haven't really needed strong mathematical (or coding) skills to create models for some time. Data Scientists add value by knowing how to translate business speak into XGBoost type model and interesting XGBoost model results into business speak. And, frankly, often by being some of the smartest people in the room. The math is occasionally helpful for speaking the language of the XGBoost model. And picking only people who are decent at math (and coding) helps ensure the smart factor. How much of that will really change with AI? I've also seen Business stakeholders try to use the chatbot to bypass the Data Scientist. Typically it's not long before there is a design decision or an interesting result the Business stakeholders don't understand. That's why I think there will be demand for Data Scientists. Not exactly evaluation and monitoring. And definitely not gatekeeping building of LLM solutions. Often the opposite, called in to explain and debug the Business stakeholders' slop.
> You recognize that you haven't really needed strong mathematical (or coding) skills to create models for some time.
And then there goes something like this [1], where researchers failed to control for p-value: "In this particular setting, emergent abilities claims are possibly infected by a failure to control for multiple comparisons. In BIG-Bench alone, there are ≥220 tasks, ∼40 metrics per task, ∼10 model families, for a total of ∼10^6 task-metric-model family triplets, meaning probability that no task-metric-model family triplet exhibits an emergent ability by random chance might be small."
Really? My instinctive reaction to this statement is that it's quackery. Would be very interested to see headlines appear about this if it's true.