ADeerAppeared

11 ms·

61.
by ADeerAppeared 2y ago
There's a fundamental problem with all these "summary" tasks, and it's obvious from the disclaimer that's on all these AI products: "AI can be wrong, you must verify this". A summary for which you must alw
62.
by ADeerAppeared 2y ago
Yes, though this scenario isn't all that interesting to discuss: * In the case of big IP holders (e.g. media companies, news organizations) this is just obtaining consent. The only fun quirck is that OpenAI's purchasing of this dr
63.
by ADeerAppeared 2y ago
I disagree, consent is important here, and the "move fast and break things" mentality is what's causing problems. #1 It's not all that certain AI training is fair use. There could be significant damages if it is found to
64.
by ADeerAppeared 2y ago
> beyond the ad-hominem Note that the "ad-hominem" pertains to only those who care about extinction risk, not AI developers in general. > which have nothing to do with scaling Except they are. There's no fundamental c
65.
by ADeerAppeared 2y ago
You're linking conservative think-tanks, not the original source from the smithsonian. If you'd look at the original infographic, it's about which norms and values are considered "default" in the US, inherited from
66.
by ADeerAppeared 2y ago
The point is that it's a separate discussion, and that losing yourself in that discussion rather than adressing the material harm of the discrimination , is a problem. And I get it, the discussion is deeply alluring. There is a deep
67.
by ADeerAppeared 2y ago
> What I am more interested in is the constant pressure on "safety risks" without anything that feels tangible to me so far. It's marketing. "Our AI is so powerful it's going to destroy the world" is just a
68.
by ADeerAppeared 2y ago
> What do you make of the placebo effect? The placebo effect is measurable . If there is no measurable improvement, there's no placebo effect either. Bear in mind that what most claims in favour of the polygraph measure is not tru
69.
by ADeerAppeared 2y ago
I am referring to their service rather than their LLM in specific. Perplexity is in the business of using an LLM to paraphrase existing content, then serving that up as their own "work" in a way that directly harms the original co
70.
by ADeerAppeared 2y ago
People aren't going to get it, because you don't get them. People have the right to learn non-copyrightable elements from your code. The claim is that AI learns copyrightable elements .
71.
by ADeerAppeared 2y ago
> Someone is likely to design an LLM that is specifically trained to do exactly that. Perplexity AI.
72.
by ADeerAppeared 2y ago
The simple version is that code is copyrightable as an expression . And the underlaying algorithm is patentable . The legal term you're looking for here is the "Abstraction-Filtration-Comparison" test; What remains if you
73.
by ADeerAppeared 2y ago
Legally? No. In spirit? ... Probably? Unlike most LLMs, Github copilot can trivially solve their copyright problem by just using only code they have the right to reproduce. They have a giant corpus of code tagged with license, SELECT BY lic
74.
by ADeerAppeared 2y ago
> If you analogy was you were a human who memorized every variation of a problem (and every other known problem) This is mere assumption. AI is supposed to work like that, but that's a goal, and not the result of current implement
75.
by ADeerAppeared 2y ago
Where it gets ethnically dubious is that: 1. The copilot team rushed to slap a copyright filter on top to keep these verbatim examples from showing up, and now claims they never happen. 2. LLMs are prone to paraphrasing. Just because you fi
76.
by ADeerAppeared 2y ago
> This is the same for humans, if we were asked to compute multiplication without thinking about it for longer than a few milliseconds. Not to be a jerk but "LLMs are just like humans when humans don't think" is perhaps no
77.
by ADeerAppeared 2y ago
I explicitly say that LLMs could do it in my response. As a show of good faith you should try reading the entire comment. Yes, I'm using simple examples to demonstrate a particular difference, because using "real" examples ma
78.
by ADeerAppeared 2y ago
The difference is what I attempt to describe at the end there. Humans apply fixed strict rules about how to break up problems, like multiplication. LLMs simply guess. That's a powerful trick to get some more capability for simple probl
79.
by ADeerAppeared 2y ago
I should clarify, "of the problem given" refers to the problem given in a prompt . As you note, transformer (and indeed, most ML) models do create a "world model". They're useful for 'specific' intelligen
80.
by ADeerAppeared 2y ago
> Doing math scales to infinity only given an error rate of zero This is true, I had omitted it for simplicity; It is still the same approach applied to scaled problems. Humans don't execute it perfectly, but computers do. With huma
81.
by ADeerAppeared 2y ago
> But correctly completing "234 * 452 =" is a lot simpler to do by doing math than by having memorized all possible multiplications. There's a fatal flaw in this theory: We can trivially test this and see that LLMs aren&#x
82.
by ADeerAppeared 2y ago
It's not a bathtub curve. Your low-level and "high"-level tasks are the same thing: Probabilistic text generation. It's not reasoning about your code, nor about the explanation it gives you. > AI has no concept of &qu
83.
by ADeerAppeared 2y ago
Yes. That is a problem . Tech execs are strongly pressured to keep the AI hype going at all costs, which just makes the damage worse when it inevitably collapses.
84.
by ADeerAppeared 2y ago
> I just don’t understand why product market fit isn’t established and revenue found before launching — these failures seem to elementary to avoid. You can ask that question of the entire tech industry. The answer is pretty straightforwa
85.
by ADeerAppeared 2y ago
No. AI firms are a good example of precisely why copyright was put in place. Right now, AI is taking people's original works and rehashing them in a way that directly competes with the original work . Some AI firms (e.g. perplexity) j
86.
by ADeerAppeared 2y ago
> why would we insert this not-even-stupid noise into the information space? Because US tech would "collapse" otherwise. Big Tech's stock price is entirely propped up by AI hype. Metaverse didn't go anywhere so they n
87.
by ADeerAppeared 2y ago
You can say that, it doesn't matter. The statistics show that these tools reduce views. And really, "I'm going to replace my entire news intake with the AI slop even if it's entirely hallucinated lies or propaganda"
88.
by ADeerAppeared 2y ago
> I think there is a real content dilemma here at work It's not really a dilemma. This is exactly what copyright serves to protect authors from. Perplexity copied the content, and in doing so directly competes with the original wo
89.
by ADeerAppeared 2y ago
> How would an LLM training on your writing reduce your reward? Because you're not getting the ad impressions anymore. The harsh reality is that people do not click on to sources , so when sites like Perplexity copy your content, y
90.
by ADeerAppeared 2y ago
Next time you "um akshually", do your homework first. > These are pretty strong guarantees, and really make it difficult for Apple to bypass. These guarantees rely entirely on trust in the hardware but it's not your hardw

More ›