Frost1x

I had a similar reaction with the language but also response times and language. It reminded me of a quote about Jon von Neumann from Edward Teller:

"Von Neumann would carry on a conversation with my 3-year-old son, and the two of them would talk as equals, and I sometimes wondered if he used the same principle when he talked to the rest of us."

I’m all for it since it’s value directly returned to humanity.

I noticed you were linking a file vs creating a correct CLAUDE.md implementation. Would you like me to fix that for you?

Marketing is just a proxy for the underlying goal: growing profit. So any evil done by marketing is driven by the pursuit of wealth. Greed underlies anything marketing does as a purer form of evil.

There’s plenty of marketing out there that just tries to make information about a product and service available without focusing on driving home higher revenue at any cost. That’s usually advertising, not marketing though, but it does exist.

So I did this yesterday for a video analysis sample with ChatGPT and it took the video, pulled out frames, did difference tests across the frames to look for significant frames to focus on, did image recognition on each frame, and interpolated motion and action between.

So I’m not sure why this says ChatGPT doesn’t “see” video and reads transcripts. Obviously if the video is already labeled that’s the shortcut. But it did an impressive job describing a video I have no inclination it would have in its training data. One could argue it wasn’t “native” and had an agent orchestrator to rely on external tools to accomplish the goal… but it worked.

Good question, I’m curious too. 911 services and cell providers come to mind, as well as subpoenaed data from law enforcement? Perhaps?

Third party commercial entities like cell providers are collecting and sharing it out of necessity but I’m guessing not selling it?

But that opens an interesting loop hole it seems where you could open a share agreement and then through other mechanisms recover the fee you’d otherwise charge for.

Provider A wants to sell data to provider B and provider B wants to buy from provider A but they legally can’t. So instead provider A just tucks the cost in some other unrelated contract with provider B with a wink wink, handshake, nod, their “relationship” then just makes them want to share the data at “no charge.” Both know the fees are tucked in other agreements, although only provider A knows the itemized cost, provider B just wonders if the cost of the other package + their friendship handshake sharing of geolocation data is worth that total cost.

To be fair, until money comes into play people tend to be less nefarious about their uses of information and intentions. Not always, but on average.

AI use is slowly creeping into pure mathematics and proving theorems or providing legging to mathematical breakthroughs. Just go watch some Terrance Tao videos to see some recent work. In addition, theorem provers and the likes have been around for awhile. Some of these systems create novel ideas or bridge novel ideas in ways that are arguably not “obvious” in any sense of the term.

While as a species our key strength has been our intelligence and it’s been core to our identity, and computing has slowly over decades infringed on this forcing us to rewrite what it is to be human, I understand the defensive view.

I also see LLMs and other AI systems spit out complete nonsense that’s truly obvious to most people. But that doesn’t make any of these systems, in my opinion, incapable of creating or bridging novel new ideas that I would call far from obvious had we substituted a human in place of it. I didn’t look at the patents in question, plenty of obvious patents make it through anymore, so that could be the case here, but I believe AI isn’t far away if not already there of creating truly patentable inventions if someone were to push it.

Then we need to redesign our entire economic system because none of it hinges on you enjoying your productivity to survive or reaping the rewards. Not saying I entirely disagree with you, just saying our economic system isn’t configured for the theoretical ideals you suggest. I’m not sure we can do that or the people with enough wealth and power to shift things would ever want to change these things.

I’m not sure all models will converge on your acceptance criteria. I’ve done quite a bit of varied agent based modeling and scientific modeling in that domain and just because you have some grounding to check against and some ideas on how you might go about getting to a convergence point doesn’t mean you’ll actually converge, you can absolutely get stuck in the information space iterating away, never finding your desired solutions.

It helps but you often have to step in the failure cases and guide them or forcibly fix certain paths to get a solution.

My opinion is that it’s a defensive mechanism. I’ve seen it across experts in knowledge domains and my own. When you hear experts disagree it’s fine because it’s another human, when the LLM disagrees and provides an objective backing that’s often solid, people jump to defense and look for very subtle nuances they wouldn’t bring up with peers and those subtleties are often highly subjective and arguably often incorrect. That’s been my observation.

I for one welcome our new LLM overlords so long as some provide be solid living standards. Mistakes do happen and they aren’t perfect so experts often do have arguments but they do come stupidly close to approximation of expertise.

I find this take off. In general, LLMs collapse a bit of all knowledge work, including many domain experts. I work with domain experts often and it’s usually the most difficult part of the process. Now, I can ask an LLM the questions I need to design the system and have an agentic system help me build parts. And it works, quite well.

The places it falls flat is domain expertise that isn’t well documented, like specific business processes. Knowing something will fail because X in dept A won’t like it and will undermine it (politics) or some silly process I didn’t know about forbids it.

So it’s not domain expertise, it’s idiosyncratic expertise that shines these days. Knowing where things deviate from a domain or the standard and being able to adapt around that. Years ago there’s a group I worked with and I was at the mercy of about 3 people I could ask questions to in order to make sure I was doing things correctly because digging into that domain was time prohibitive. Now, I can sit down one evening and dig fairly deep into a domain. I can understand common practice, approach, acronyms, nomenclature, so on. I can find other popular competing systems. I can rapidly figure out what I need.

The same is true for software, agents don’t collapse software they help people build fairly usable systems but they don’t find the expertise a senior level engineer has where designs or implementations may fail in practice. The idiosyncrasies of software and software systems, especially in your very specific set of constraints you need to operate in that may deviate from the rest of the world.

The results are the same but I’ve found the process to get to the results are just more pleasant with Claude. I can’t put my finger on it. Overall most these models at the highest level are about the same in many respects but the UI/UX for some are just more enjoyable, for lack of a better term.

Codex I feel the need to be very specific and precise with. Claude… I feel like I can be lazy, which I enjoy.

Both still need to be reviewed stringently but I feel I can be more ambiguous with Claude and get better results than when Codex.

Maybe money but clearly not power. The American voting population gave him more power than pretty much anyone in many senses of the word “power,” yet he’s clearly not loyal to the American voting population.

So I think money or wealth is the bigger weight here.

Is it though? You assume the abstractions in Python are battle tested and you understand them. Usually people are relying on arbitrary libraries so unless you’re constraining libraries and those libraries have good review processes, it won’t be long until high level functions you’re reading are generated by LLMs to, so to review your LLMs use of other LLM generated functions you have to drop down a few levels and review at that level.

At some point that becomes less sustainable and looking at something with less abstraction assures you’re at least looking at a baseline source of truth, even if the volume is massive.

There’s going to be a whole world in the knowledge economy, not just software but everywhere, around validation and sign off of information that we’ve taken for granted as a cost prohibitive process where only the best options make it to high levels of function and maturity.

When you get some abstraction working you concretize it in something deterministic, or sort of “cache” that knowledge bit (aka write me a function, class, library, whatever). In the future, the nondeterministic path now has a deterministic piece to lean on as it explores the problem space. Rinse, repeat, eventually you have a mostly deterministic system now. Leave flexibility in space where you need that nondeterminism.

I mean, the ad doesn’t necessarily have to be made aware of the exact prompt context, just that the ad itself was relevant. You can basically have the ads prequalified for areas and serve them when relevant. Now that does show the user is talking about something relevant most likely, and depending on how they decide to serve them or provide referring, it may traceable to a profile/identity built for that user externally.

I’d be more concerned as to how this ends up in agent platforms using the LLMs, when you don’t have a fairly autonomous agent based system using these the entire point is that a human isn’t involved, so who are you serving ads to and where are you injecting them.

Moreover, if you are injecting them everywhere, does that survive stare for subsequent steps, meaning from the first set of results I get, does that loop back in again with the ad injected into the context. Because now, we have yet another dangerous way of injecting instructions into an already issue prone surface area.

I’m guessing they’re going to have special APIs that don’t include ads, and those are going to cost more, especially for non embedded agents (processes that already exist inside ChatGPT that kick off transparently from prompts, like asking it to work with an office document). After all the customers using agents aside from developers are mostly businesses, so it’s where the money is. The ads will exist for the poor to subsidize their use, and probably create even more barriers for agentic use like I described. Just my thoughts.

And good luck litigating against any business in this administration. Unless they explicitly tick off certain people or refuse to kiss the ring, they can get away with almost anything right now and there’s little risk of doing it or not because ticking off this admin will raise illegitimate prosecution even if you’re perfectly legal, almost the same level of if you’re not. It’s the ideal playground for doing all sorts of manipulation, just kiss the ring and you’ll be fine.

Why does it need to be revenge porn? Pretty sure regular old porn has a large market there where people can specify what they idealistically want to see vs trying to find it, if it exists.

Not every place has LEGO incest porn… or whatever the kids are into these days.

It’s probably more similar to Japan in terms of cultural tolerance. I heard the same story years ago and only recently visited (just after the Paris Olympics). I usually try to learn some of the basics of the language before visiting but was incredibly busy and didn’t this trip. I had no issues and I was all over Paris. People were very reasonable, and translation apps/services helped me plenty, but for the most part they spoke English or could understand some basic level of it. If you live there and try to assimilate but speak poorly or little, there may be less tolerance? As a tourist I had not a single incident.

I don’t like to be the ugly American who just assumes the world should speak my language, so I was ready for language barriers, but I had no real issues at all.

It’s interesting because I’m seeing some emerging conversations where users are tending to prefer general agents that have their preferential bias over more constrained or specially built agents, because there are certain arbitrary goal criteria they either have forced on them or want to force upon the agent and the general purpose agents tend to do well at this because they just trudge along and do whatever.

Meanwhile more specialized agents that try to add or enforce constraints around a problem space where certain aspects tend to be well established don’t sit well with a lot of uses. “No, you and general knowledge don’t know best, I know best… do this.”

I can see the use case for both but I’m seeing a whole lot more willingness to want confirmation bias, essentially to automate away parts of jobs and tasks people already do but in the personalized or opinionated way they’ve established, unwilling to explore alternative options.

So the general purpose agent structures that just kickoff whatever they can tend to favor best in terms of positive feedback from agent users. Meanwhile it to some degree ignores many of the potential benenfits of having agents with general knowledge and bounded by general established bounds. It’s basically the whole “please do parts of my job for me but only the way I want them done.”

People aren’t ready for being wrong or change, they just want to automate parts of their processes away. So I’m not sure “no” is going to sit well with a lot of people.

Not nearly your age but I agree with your sentiments entirely. I mainly focused on using computing not for business purposes but scientific purposes and how we can forward science using compute and technology and I’ve felt much the same way for some time. The new layers and layers of abstraction added little in the way of productivity to getting to the root problems I wanted to and there have always only been so many hours in the day and dollars in the sponsoring agency’s purse to pursue new innovative work.

Now a lot can be cast off to LLMs to focus on the problem space and the innovative computing use around them. It’s been exciting to not worry about arbitrary idiosyncrasies and machete through jungles of technical minutia to get to the clearing. I still have to deal with them but less of them. And I don’t have to commit nearly as much in the technical space to memory to address problems, I can often focus on higher level architectural decisions or new approaches to problems. It’s been quite enjoyable as well.

I don’t think this is a good long term solution. LLMs can do easy language substitutions and you can even force them to add errors. So relying on that alone won’t work as people intentionally make things look more “human.”

I think it’s taken by greed focused extremists, they’re just trying to bide favor with some other extremist groups as their flail to maintain their power and attempt to expand it.

In the security space you’re encouraged to be as transparent as possible. Most modern forms have ample space to write in detailed explanations.

I have some silly not nearly as interesting infractions and I wrote them out in detail explaining, without any issue in processing background checks. It usually is something that’s asked about in an in person interview at that point.

Tech has slowly been moving that way anyways. In terms of ROI, you’re often much better off targeting whales and large clients than trying to become the ubiquitous market service for consumers. Competition is fierce and people are poor comparatively, so you need the volume for success.

Meanwhile if you go fishing for niche whales, there’s less competition and much higher ROI for them buying. That’s why a lot of tech isn’t really consumer friendly, because it’s not really targeting consumers, it’s targeting other groups that extract wealth from consumers in other ways. You’re selling it to grocery stores because people need to eat and they have the revenue to pay you, and see the proposition of dynamic pricing on consumers and all sorts of other things. Youre marketing it for analyzing communications of civilians for prying governments that want more control. You’re selling it to employers who want to minimize labor costs and maximize revenue, because they have millions or billions often and small industry monopolies exist all around, just find your niche whales to go hunting for.

And right now I’d say a lot of people in tech are happy to implement these things but at some point it’s going to bite you too. You may be helping dynamic pricing for Kroger because you shop at Aldi but at some point all of this will effect you as well, because you’re also a laboring consumer.

We really have a societal problem in that we allow private entities to do things we don’t allow government to do. Furthermore, the issue is exacerbated by then allowing governments to bypass these issues by then just paying private entities to do the things it can’t do as a proxy for the same functional outcomes.

But we want to support privatization at all cost, even when privatization these days has significant influence on our daily lives, akin to the concerns we had when we placed restrictions on government. Seems like we need to start regulating private actions a bit more, especially when private entities accumulate enough wealth they can act like multi state governments in levels of influence. That’s my opinion, at least.

I think you’re missing the main limiting resource: money.

Some of these projects could occupy entire regions of cloud compute in some cases for awhile, some even more depending on the problem. But running that for even a short time or decades needed would cost more money than anyone has to do.

Academic HPCs existed long before cloud compute options and for certain problem spaces could also be used even in non-distributed memory cases to handle this stuff. But you still needed allocation time and sometimes even funding to use them, competing against other cases like drug design, cancer research, nuclear testing… whatever. So searching for ET could be crowdsourced and the cost distributed which is something that made it alluring and tractable.

I used to run a small academic cluster that was underutilized but essentially fully paid for. I’d often put some of these projects running as background throttled processes outside scheduler space so the 90% of the time no one was using them, the hardware would at least be doing some useful scientific research since it’s after-all funded largely from federal scientific research funding. There was of course some bias introduced by which projects I chose to support whereas someone else may have made a more equitable choice.

Complain to our representatives who will do absolutely nothing because the system is ripe for abuse and we’ve put people who actively want to abuse and exploit it into office.

I keep telling everyone and have been for a year, it’s not just our problem, due to global US positioning it’s now a world problem. Just ask Venezuela. Regardless of what you think about the end result the ends did not justify the means.

I for one will be collecting my (completely legal) hunting rifles and weapons I’ve had in storage since I was a kid, have them professionally serviced and grab some ammunition, on the terrible case I need to defend myself which I thought I’d never ever have to consider and I’d just sell them some day. But alas we have a lot of really really stupid as well as downright toxic voters in this country.

In my opinion it’s to some degree an artifact of immature and/or rapidly changing technology. Basically not many know what the best approach is, all the use cases aren’t well understood, and things are changing so rapidly they’re basically just creating interfaces around everything so you can change flow in and out of LLMs any way you may desire.

Some paths are emerging popular, but in a lot of cases we’re still not sure even these are the long term paths that will remain. It doesn’t help that there’s not a good taxonomy (that I’m aware of) to define and organize the different approaches out there. “Agent” for example is a highly overloaded term that means a lot of things and even in this space, agents mean different things to different groups.

Eh, I like an interesting spin on a classic. I’ve seen/heard the Frankenstein plot and small variations on it many times, taking a different direction is a good way to keep in a general universe but develop something new. If you’re not going to come up with new interesting content, at least don’t rehash the exact story I’ve heard many times. But that’s just my preference—I really enjoyed it and have become a fan of Guillermo del Toro works recently (due to exposure on Netflix). I’m not huge critic really so I won’t speak to artistic merit but I can at least say I really enjoyed it.

I would argue not good enough but better. A home cinema depending on viewing distance can have superb visual qualify. Comfort is going to be impossible to beat to being at home. A lot of theater projectors top out at 4k just like home TVs and they’re not as bright. Also information density is lower (it’s 4k spread over a huge wall).

The only shortcoming now really is if you want to view with several people and socialize after, it may be difficult for someone to accommodate a large party with good viewing in their home without a theater setup. And of course audio, audio is where theaters can still stand out. It’s a pain in the ass for most homes to setup a good sound system, you really often do want a dedicated theater area which most aren’t going to have. A soundbar helps. You can Jerry rig some surround speakers into any space but it’s often a pain. So that’s really the last barrier: cheap low latency sound that can beat a theater.

For me comfort trumps the slightly degraded sound. Plus some baby crying or random person chatting during the movie can break that as well.