Ready at what level, though. The subtleties are what matters.
It’s well established that belligerents can use mines, to separate the tactical decision of deploying for purposes of area denial; from the snap-second lethal decision (if one can stretch that definition) to detonate in response to an triggering event.
Dario’s model prohibits using AI to decide between enemy combatant and an innocent civilian (even if the AI is bad at it, it is better than just detonating anyways); Sam’s model inherits the notion that the „responsible human” is one that decided to mine that bridge; and AI can make the kill decision.
How is that fundamentally different in the future war where an officer might make a decision to send a bunch of drones up; but the drones themselves take on the lethal choice of enemy/ally/no-combatant engagement without any human in the loop? ELI5 why we can’t view these as smarter mines?
It's different because we are talking about a technology that we might lose control over at some point. Those drones in your example might make an entirely different choice than what you anticipated when you let them take off.
Well, or just the possibility of future-proofing the agreement in favor of the US government, as well as walking back the slippery slope of „no autonomic lethality” and „no mass surveillance”.
The former, grants the Congress the ability to change the definition of all „lawful use” as democratically mandated (if the war is officially declared, if the martial law is officially declared).
The latter, is subtle. There can exist a human responsibility for lethal actions taken by fully autonomous AI - the individual who deploys it, for instance, can be made responsible for the consequences even if each individual „pulling of a trigger” has no human in the loop (Dario’s PoV unacceptable).
Similarly, and less subtly, acceptance of foreign mass surveillance, domestic surveillance (as long as its lawful and not meeting the unlawful mass surveillance limits!) seems to be more in the Pentagon’s favor.
Whether we like it or not, we’re heading into some very unstable time. Anthropic wanted to anchor its performance to stable (maybe stale) social norms, Pentagon wanted to rely on AI provider even as we change those norms.
Not the person you're replying to, and I may be wrong about this, but Amazon?
Jeff's original vision was "relentless customer focus" and ...
actually on second thought I'm seeing the argument 'Amazon stopped caring about customers and is in full enshittification mode at this point'.
But maybe Amazon circa ~2010/2015, or Google around 2010 was still pretty close to the original vision of customer service/organizing the world's information.
Or Apple? They're still making nice computers, although not sure they count as VC backed.
Well Google‘s vision was to catalog all the world’s data
Apple wanted to make personal computing stable - they were absolutely VC backed
I suppose the original question is vague enough that it could always encompass everything which is founders vision even if the vision changes so it’s like OK well then then there’s nothing really to say that you’re stable too it’s just some whatever the function of the person who started the organization is and even that you could debate
FWIW, I don’t actually know if board of Anthropic has actual power to replace its CEO or if Dario has retained some form of personal super-control shares Zuckerberg style.
At some level of growth, the dynamics between competent founders and shareholders flip. Even if the board could afford to replace a CEO, it might not be worth it.
I'd counter that at this level of capital, if the CEO doesn't well align with the capital, then super-control shares will be overpowered by super-lawyers and if there is need some super-donations. OpenAI was a public interest company...
Not at all. Especially at that level of capital. It’s the equity equivalent of „if you owe a bank a million dollars, you’re in trouble. If you owe a bank a billion dollars, the bank is in trouble”.
Capital is extremely fungible. Typically extremely overleveraged. Lawyers are on the other hand extremely overprotective. They won’t generally risk the destruction of capital, even in slam-dunk cases. Vide WeWork.
Anthropic has an odd voting structure. While the CEO Dario Amodei holds no super-voting shares, there are special shares controlled by a separate council of trustees who aren't answerable to investors and who have the power to replace the Board. So in practice it comes down to personal relationships.
Except they did with Wine, in a way. They got to the point where sufficient number of third party software developers target the common base between Wine and Windows (Steam/Proton), electing to have broader compatibility rather than catching all the newest Windows-only APIs.
I wonder how much similar behavior influence other buying choices. I’ve been eyeing an upgrade from M1 for a while - so far punting on it, mostly because of Asahi.
I guess I wasn't aware that Wine pivoted from trying to be a general purpose, drop-in replacement for Windows to being a platform for games that only supports a subset of Windows functionality.
It's much more difficult to keep current and support the full functionality of a much larger competitor's offering when you have to support everything. In my experience it was an all or nothing proposition. Either you emulated it 100% or you had nothing. I think Asahi is more in this realm maybe than Wine. It really needs to support all the hardware, 100%, or it's value is greatly diminished.
> I guess I wasn't aware that Wine pivoted from trying to be a general purpose, drop-in replacement for Windows to being a platform for games that only supports a subset of Windows functionality.
Or „just enough” for the subset of users that is „enough” to ensure product viability. The absolutism of „all or nothing” is rooted in the strictly-better mentality for replacing something.
For Wine/Proton, the core demographic is essentially gamers, who tend to overlap heavily with engineering population later on, and thus core population for Microsoft to capture and retain. Once Steam removed that vendor lock-in, the corporate discussion became more flexible.
For Asahi (proud Asahi user for 4y now), the added value of „most powerful Linux/Arm64 laptop on the market” outweighs the few things that don’t work on Asahi (HDMI out is probably the only one that occasionally matters for me, but screencasting works well enough). Yes, there are gaps, but they are smaller than things from Linux that are missing on OSX or Windows for me.
There was a PR for Adobe products a few weeks ago (https://github.com/ValveSoftware/wine/pull/310), though it seems like they're redirecting it to the main Wine repo now since it makes more sense there
I think people (especially those who joined the internet after the .com bubble) underestimate the level of decentralization and federation coming with the old-school (pre web-centric mainframe-like client mentality) protocols such as email and Usenet and maybe even IRC.
Give me “email” PR process anytime. Can review on a flight. Offline. Distraction free. On my federated email server and have it work with your federated email server.
And the clients were pretty decent, at running locally. And it still works great for established projects like Linux Kernel etc.
It’s just pain to set up for a new project, compared to pushing to some forge. But not impossible. Return the intentionality of email. With powerful clients doing threading, sorting, syncing etc, locally.
I'm older than the web. I worked on projects using CVS, SVN, mercurial, git-and-email, git-with-shared-repository, and git-with-forges. I'll take forges every time, and it isn't even close. It's not a matter of not having done it the old way, it's a matter of not wanting to do it again.
I guess we might have opposite experiences. Part of which I understand - the society moved on, the modern ways are more mature and developed… but I wonder how much of that can be backported without handing over to the centralized systems again.
The advantage of old-school was partially that the user agents, were in fact user agents. Greasemonkey tried to bridge the gap a bit, but the Web does not lend itself to much user-side customization, the protocol is too low level, too generic, offering a lot of creative space to website creators, but making it harder to customize those creations to user’s wants.
I'm older than the trees, but, younger than the mountains! Email all day, all the way. Young people are very fascinated and impressed by how much more I can achieve, faster, with email, compared with their chats, web 3.0 web interfaces, and other crap.
Yes, it takes time to learn, but that is true for anything worthwhile.
What I like about git-and-email-patches is the barrier to entry.
I think it's dwm that explicitly advertises a small and elitist userbase as a feature/design goal. I feel like mailing lists as a workflow serve a similar purpose, even if unintentionally.
With the advent of AI slop as pull request I think I'm gravitating to platforms with a higher barrier to entry, not lower.
Another reason why I think it failed (TIL Yann LeCun was the coauthor) is the connotation with the pirate books/articles community.
When I came across this format in college days, when handling lots of scanned material, it always triggered the mental “don’t install suspicious software” block. Which is a shame as the article points out it was the superior format.
The better question is the seek latency. The bandwidth for read isn’t too horrible, if the seeks can be kept within reason. This is somewhere between tapes and actual pressed optical media (not dyed /re/writeable). Should seek way faster than tape (maybe even on par with BluRay) and 30Mbps read is manageable for doomsday scenarios.
Long term databanks. Libraries. GitHub’s archive bunkers. Microfilm replacements.
at theoretical perspective, if the X-Y plane can be addressed with 16-bit DAC by controlling laser deflection. then to seek any data with in a 4GB address space will have typical latency of 300us with the latest laser scanning technology.
I am not aware any laser scanning technology that can do 16-bit accuracy that has no moving part. so, fundamentally, this is a storage technology with mechanical addressing.
laser can be scanned by acoustic wave, but that itself lack the beam pointing accuracy. the ultrasonic drive frequency will limit how fast is can deflects the laser beam.
Essentially, yes, they haven’t done deep software. Netflix probably comes closest amongst FAANG.
Google, Meta, Amazon do “shallow and broad” software. They are quite fast at capturing new markets swiftly, they frequently repackage OpenSource core and add the large amount of business logic to make it work, but essentially follow the market cycles - they hire and layoff on a few year cycle, and the people who work there typically also will jump around industries due to both transferable skills and relatively competitive competitors.
NVDA is roughly in the same bucket as HFT vendors. They retain talent on a 5-10y timescales. They build software stacks that range from complex kernel drivers and hardware simulators all the way to optimizing compilers and acceleration libraries.
This means they can build more integrated, more optimal and more coherent solutions. Just like Tesla can build a more integrated vehicle than Ford.
I have deep respect for cuda and Nvidia engineering. However, the arguments above seem to totally ignore Google Search indexing and query software stack. They are the king of distributed software and also hardware that scales. That is way TPUs are a thing now and they can compute with Nvidia where AMD failed. Distributed software is the bread and butter of Google with their multi-decade investment from day zero out of necessity. When you have to update an index of an evolving set of billions of documents daily and do that online while keeping subsecond query capability across the globe, that should teach you a few things about deep software stacks.
These companies innovate in all of those areas and direct those resources towards building hyper-scale custom infrastructure, including CPU, TPU, GPU, and custom networking hardware for the largest cloud systems, and conduct research and development on new compilers and operating system components to exploit them.
They're building it for themselves and employ world-class experts across the entire stack.
How can NVIDIA develop "more integrated" solutions when they are primarily building for these companies, as well as many others?
Examples of these companies doing things you mention as being somehow unique to or characteristic of NVIDIA:
You're suggesting Waymo isn't deep software? Or Tensorflow? Or Android? The Go programming language? Or MapReduce, AlphaGo, Kubernetes, the transformer, Chrome/Chromium or Gvisor?
You must have an amazing CV to think these are shallow projects.
No, I just realize these for what they are - reasonable projects at the exploitation (rather than exploration) stage of any industry.
I’d say I have an average CV in the EECS world, but also relatively humble perspective of what is and isn’t bleeding edge. And as the industry expands, the volume „inside” the bleeding edge is exploitation, while the surface is the exploration.
Waymo? Maybe; but that’s acquisition and they haven’t done much deep work since. Tensorflow is a handy and very useful DSL, but one that is shallow (builds heavily on CUDA and TPUs etc); Android is another acquisition, and rather incremental growth since; Go is a nth C-like language (so neither Dennis Richie nor Bjarne Stroustrup level work); MapReduce is a darn common concept in HPC (SGI had libraries for it in the 1990s) and implementation was pretty average. AlphaGo - another acquisition, and not much deep work since; Kubernetes is a layer over Linux Namespaces to solve - well - shallow and broad problems; Chrome/Chromium is the 4th major browser that reached dominance and essentially anyone with a 1B to spare can build one.. gVisor is another thin, shallow layer.
What I mean by deep software, is a product that requires 5-10y of work before it is useful, that touches multiple layers of software stack (ideally all from hardware to application) etc. But these types of jobs are relatively rare in the 2020s software world (pretty common in robotics and new space) - they were common in the 1990s where I got my calibration values ;) Netscape and Palm Pilot was a „whoa”. Chromium and Android are evolutions.
> No, I just realize these for what they are - reasonable projects at the exploitation (rather than exploration) stage of any industry.
I get that bashing on Google is fun, but TensorFlow was the FIRST modern end-user ML library. JAX, an optimizing backend for it, is in its own league even today. The damn thing is almost ten years old already!
Waymo is literally the only truly publicly available robotaxi company. I don't know where you get the idea that it's an acquisition; it's the spun-off incarnation of the Google self-driving car project that for years was the butt of "haha, software engineers think they're real engineers" jokes. Again, more than a decade of development on this.
Kubernetes is a refinement of Borg, which Google was using to do containerized workloads all the way back in 2003! How's that not a deep project?
True, for some definition of first and some definition of modern. I’d say it builds extremely heavily on the works inside XTX (and prior to that, XFactor etc) on general purpose linear algebra tooling, and still doesn’t change the fact that it remains shallow, even including JAX. Google TPUs change this equation a bit, as they are starting to come to fruition; but for them to reach the level of depth of NVDA, or even DEC to SUN, they’d have to actually own it from silicon to apps… and they eventually might. But the bulk of work at Google is narrow end-user projects, and they don’t have (at large) a deep engineering excellence focus.
Waymo is an acquihire from ‘05 DARPA challenges, and I’d say Tesla got there too (but with a much stricter hardware to user stack, which ought to bear fruits)
I’d say Kubernetes would be impressive compared to 1970s mainframes ;) Jokes aside, it’s a neat tool to use crappy PCs as server farms, which was sort of Google’s big insight in 2000s when everyone was buying Sun and dying with it, but that makes it not deep, at least not within Google itself.
But this may change. I think Brin recognizes this during the Code Red, and they start very heavily on building a technical moat since OpenAI was the first credible threat to the user behavior moat.
You think that Tesla, which has not accepted liability for a single driverless ride, has "gotten there?" I'm not even going to look up how many Waymo does in a month, I'm sure it's in the millions now.
Come on, man.
> Google's TPUs change this equation a bit
Google has been using TPUs to serve billions of customers for a decade. They were doing it at that scale before anyone else. They use them for training, too. I don't know why you say they don't own the stack "from silicon to apps" because THEY DO. Their kernels on their silicon to serve their apps. Their supply chain starts at TSMC or some third-party fab, exactly like NVIDIA.
Google's technical moat is a hundred miles deep, regardless of how dysfunctional it might look from the outside.
Yeah I'm not convinced Microsoft can do software anymore. I think they're a shambling mess of a zombie software company with enough market entropy to keep going for a long time.
They did engineer a good browser: original Edge with the Chakra JavaScript Engine. It was faster than Google Chrome and had some unique features: a world-best, butter-smooth and customizable epub reader. I loved it for reading - it beat commercial epub readers - and then Nadella took over and said Microsoft is getting rid of it and Edge will move to Chromium and Microsoft will also get rid of Windows phone. Modern Microsoft will be Cloud/AI and Ads. That was so depressing.
You could say the same thing about all Microsoft products then. How many full time developers does it take to support Windows 11 when Linux is available, SqlServer when Postgres is available, Office when LibreOffice exists?
And so on all under licenses that allows Microsoft do whatever it wants with?
They should be embarrassed to do better, not spin it into a “wise business move” aka transfer that money into executive bonuses.
Microsoft gets a lot of its revenue from the sale of licenses and subscriptions for Windows and Office. An unreliable source that gives fast answers to questions tells me that the segments responsible for those two softwares have revenue of about $13 and about 20 billion per quarter respectively.
In contrast, basically no one derives any significant revenue from the sale of licenses or subscriptions for web browsers. As long as Microsoft can modify Chromium to have Microsoft's branding, to nag the user into using Microsoft Copilot and to direct search queries to Bing instead of Google Search, why should Microsoft care about web browsers?
It gets worse. Any browser Microsoft offers needs to work well on almost any web site. These web sites (of which there are 100s of 1000s) in turn are maintained by developers (hi, web devs!) that tend to be eager to embrace any new technology Google puts into Chrome, with the result that Microsoft must responding by putting the same technological capabilities into its own web browser. Note that the same does not hold for Windows: there is no competitor to Microsoft offering a competitor to Windows that is constantly inducing the maintainers of Windows applications to embrace new technologies, requiring Microsoft to incur the expense of applying engineering pressure to Windows to keep up. This suggests to me that maintaining Windows is actually significantly cheaper than it would be to maintain an independent mainstream browser. An independent mainstream browser is probably the most expensive category of software to create and to maintain excepting only foundational AI models.
"Independent" here means "not a fork of Chromium or Firefox". "Mainstream" means "capable of correctly rendering the vast majority of web sites a typical person might want to visit".
They did incur that cost… for decades. They were in a position where their customers were literally forced to use their product and they still couldn’t create something people wanted to use.
It’s well established that belligerents can use mines, to separate the tactical decision of deploying for purposes of area denial; from the snap-second lethal decision (if one can stretch that definition) to detonate in response to an triggering event.
Dario’s model prohibits using AI to decide between enemy combatant and an innocent civilian (even if the AI is bad at it, it is better than just detonating anyways); Sam’s model inherits the notion that the „responsible human” is one that decided to mine that bridge; and AI can make the kill decision.
How is that fundamentally different in the future war where an officer might make a decision to send a bunch of drones up; but the drones themselves take on the lethal choice of enemy/ally/no-combatant engagement without any human in the loop? ELI5 why we can’t view these as smarter mines?
reply