Yeah, we'll see how long that lasts.
I'd bet 500 thousand dollars that this sentiment goes away as fast as they were rerunning cold and flu medicine commercials after everyone was sick of giving a shit about COVID.
836 karma · joined November 29, 2020
I'd bet 500 thousand dollars that this sentiment goes away as fast as they were rerunning cold and flu medicine commercials after everyone was sick of giving a shit about COVID.
- a person demoing something they made
- "we should say this is authored by Claude."
- "we should say this was a human directing Claude"It's been 9 years since I've quit smoking and still to this day out of nowhere I will have a thought, "ok, quick smoke and on to the next task"
Recently it happened to me and I was wondering if there are just some neural pathways that are permanently wired to that reward function
The prevailing moral panic prevents anyone in the space from sharing in communities where open source is generally discussed. People flag and openly disparage (against community guidelines) and have histrionic reactions.
We'll just have to wait until the topic cools off before we can see how llm usage impacts open source.
As shitty as this is, I'm glad that it's happening because things like this erode the shamers credibility.
There were plenty of histrionic COVID warriors that tried to ruin lives and businesses. I bet the venn diagram of those people is close to a circle.
On optimistic days, I'd say, "it is solved". But it's a little more nuanced.
For example, My pipeline takes machine code into "readable" c that matches the underlying binary function by function, and then has an agent annotate each function with what it thinks it should do.
What that produces is decent, but it needs more refinement as more of the project evolves and context emerges.
The problem is that there's an almost infinite source space between what can be statically decompiled and matching c code with the appropriate idioms. If the model can't reason about the flavor of c code it should produce, or doesn't have adequate tooling to unwind a particular function it will just grind away and eventually give up. Sometimes a single function can take a lot of time. The main loop of the game that's being decompiled right now took over a day and multiple sessions and eventually needed to be escalated to fable.
I'm still working on tools and the harness to guide the llm better. I'm optimistic that this can be made more efficient, but still playing with it.
Maybe I'll buy some run down property and fix it up. Try to make some money consulting, build some games.
I already live below my means and have a decent amount of money saved up, but still feel like it isn't enough to just stop working.
Right now, there is a special cadre of people who think they can stop what's currently happening with llms by being abusive and extremely negative. And for some reason, the abuse is just totally accepted in communities that have historically touted themselves as inclusive.
These folks are borderline mentally ill, or at least exhibit extreme and over the top reactions that are symptomatic of mental distress. Feel empowered to ignore them completely. Take it as a signal that you're in the right space. Nothing that ever mattered was extremely popular at first. Challenging the status quo is painful.
The vocal minority is always loudest on platforms that give them a voice. I'll share some anecdata with you: I'm in an IC leadership position in big tech, I see what's happening in a lot of the top tech companies. In the "real world" (bay area, fortune 10, 50, 100) these companies are all-in on things like what you're showing here. It will never go back to the way things were.
Your project serves as part of interesting case studies for what llms are capable of. Only you can give an honest assessment to your question:
Given what you've seen with this project, do you think that coding services are a failure? What did you learn?
Then don't be that guy.
> But...
Oh, ok.
Hackernews, dang and the other mod are constantly telling people, wE'rE dOIng sOMetHinG dIffeReNt hERe
Yeah, complaining about fucking ai instead of discussing the topic.
I don't give a fuck about how whatever it is you perceive as "claudisms" makes you feel.
The author is deeply technical, and the writeup as presented is easy to follow.
If you couldn't follow it and you are technical, then you have a form ai psychosis: the kind that makes you so hyper-viliglent to seeing something that might have been produced in a way that you don't like that it blanks out your brain.
If you need religion, I'm not going to stop you.
Also, you're condescending tone at the end is adorable. Very cute. Mommmmy nooo you got meee. I'm sowwwyyyy
Effective Altruism is in of itself a philosophy developed by charlatans that refuse to engage in the tradition that preceded it by about 2500 years.
A disclosure should be required next to anything they're working on.
Nothing will save us. You can't automate trust.
That's so cool and insightful.
No, just stop having pets.
I don't think you're a crazy vegan, and I'm not outraged. But your proposal is ridicules.
1 - https://www.theguardian.com/sustainable-business/2017/jul/10...
2 - https://influencemap.org/briefing/Carbon-Majors-2024-Data- Update-35466
3 - https://www.theenergymix.com/17-of-worlds-20-biggest-carbon-...
4 - https://www.rinnovabili.net/environment/pollution/world-co2-...
- people who like writing code for codes sake
- people who like writing code to build things
In the past, these folks have lived more or less in harmony unified under the common interest of applied maths, engineering and architecture that is necessary to write things that require a lot of vode. Aside for a few classic and very public spats.
Now that community is fractured because llms are increasingly making writing code irrelevant. If that weren't true, we wouldn't be having this convo at all.
I think it's perfectly reasonable to want to cultivate spaces that preserve that craft, but wish they wouldn't be so obnoxious about it. I imagine that a good handful of these folks in the "no llm ever camp" collectively rolled their eyes at vegans and animal rights advocates at some point, and are now feeling being on the unpopular side of progress.
Personally I think it's a cool prospect to have Trappist Monk-like communities producing hand crafted code. It will probably be really good and sturdy. We'll look at it in 10 years and say, "damn they really don't make it like that anymore!"
Having said that, consider taking more care when authoring articles with the help of llms.
There's a lot of nonsense in this piece that distracts from the overall point you're trying to make. Things such as
- "the output price is frozen at the anchor
- " This post is that gap collapsing to zero and the discovery that the real fight was never about price."
- "Strip away everything and the economics reduce to a single line"
- "Feed that into the equation and unit cost is a hyperbola in utilization"
- "the plot twist"
- "Two data points deserve a hard stare."
- "Caveat, because the numbers deserve it."
- "Sizing the prize: what does one box actually hold?"
And so on..
Nobody likes this, and it diminishes your credibility as an author. Llms output stuff like this because they are confabulating prose between facts. It spaces out arguments and adds text, but it's just meaningless nonsense that distracts from the overall point.
It sounds like you did a real thing here and that you might have some interesting results to share. If you presented them clearly in your own voice, more people will be interested.
We only have a limited sample size of economic data.
No, we don't. It's quite good, but it's nowhere near perfect. For example, you can see: https://genai-showdown.specr.net/ or others that highlight how far from perfection we still are.
> and it hasn't really changed the nature of human expression.
It sure has changed our discourse. Look at hackernews, where people just can't help themselves to engage in ragebait and flamewar comment threads about llm generated accusations. We've got half the people convinced that looking at ai content is like consuming food.
I'd venture a guess that it's too early to determine how this technology will change expression. A good analog would be photography, which caused a similar meltdown in the arts at the time.
> I don't see my friends getting wildly creative.
Maybe you don't have creative (enough) friends.
The majority of top threads on hn on any given days are legit flame wars and llm accusation mud slinging.
But you cant say Dario Amodei is an effective altruist. Incredible.
Posted by the account, BurningFrog. You can't make this shit up.
Forgive me if I do not take "making sunsets" as a credible source of information.
There's a body of research that covers existential risks associated with sai and srm:
- https://www.sciencedirect.com/science/article/abs/pii/S00163...
- https://www.frontiersin.org/journals/climate/articles/10.338...
- https://agupubs.onlinelibrary.wiley.com/doi/10.1002/2017EF00...
Further from an actual credible source:
https://www.unep.org/resources/report/Solar-Radiation-Modifi...
> While there is a growing body of literature and advanced modelling on the topic, which has been developing for many years, there is very little evidence or research on the risks and impacts of SRM. The report outlines these limitations in the evidence base. UNEP is concerned that while the scientific literature shows that SRM, and stratospheric aerosol injection specifically, can cool the globe, there is a dearth of data on the risks and impacts. For technologies that can alter the chemistry of the stratosphere, the world needs much more information to support decision-making.
No, it is not "safe" by any measure.
You're frustrated because you think you understand a thing that you have only a tenuous grasp on, from the deep expertise of a single area that is not transferable to whole of the problem. You then transfer that frustration on to "people like me" whom you seem to think are doing something bad. But really, the commentary is that the claims you put forward are insubstantial and lack credibility.
The sum total of what you've put forward here and in other threads amount to arguments of authority based on the musings of a handful of capitalists. Then cite a body of thoughts from a group of transhumanists who think they are smart enough to reinvent areas philosophy without engaging in the body of work that predates the movement by 2000+ years.
Folks who have been involved with lesswrong have made good advancements in the domain of machine learning, and that's where it stops. They're not some collection of prescient macroeconomic and geopolitical savants. They're also wrong about as much as they are right (cryonics), and produce plenty of whiffs (sbf, zizians). You're suggesting that we listen to a broken clock because it's been right once.
Your dismissal of climate (read: climate breakdown) as "something less severe" is truly demonstrative of the cognitive bias that rips through these groups. You're so frustrated that people like me won't listen to the "experts" in one breath, and in the next diminish the 50 years of actual Hard Science that we have that shows how severe the future of climate breakdown will be. Go read the latest ipcc report, the most conservative scientific organization on earth is sounding the alarm bells.
Now we get back to this openai incident. We're watching Edison electrocute Topsy on Coney Island. There's a group of people saying, "this is obviously a publicity stunt". And there's a group of people saying, "the elephant is actually dying obviously it's real."
The fact of the matter is that, yes electricity is powerful, but Edison is electrocuting the elephant. At that moment, there is no body of evidence that shows with any certainty how electricity is going impact society. And there's certainly no compelling arguments that only Edison should be able to decide how to use it. He's electrocuting a fucking elephant.