I remember when there was a kid who kept installing this and the Chex doom on the school pcs in 7th grade. GTA was pretty controversial as a game even though it is incredibly tame by today’s standards. I’m pretty sure they never caught him.
I got in trouble for bringing Doom shareware and putting it on one of the computers in the lab at school... worth, though - had half the class exclaiming at the awesomeness until we got busted hahaha
20 juniors become some % of 20 seniors. and some % of that principals. Even if it lives up to the claims you’re still destroying the pipeline for creating experienced people. It is incredibly short sighted.
The thing I really love about magic is that so much of it boils down to “I practiced for years and developed a seemingly superhuman ability to manipulate this object”.
I really relate to this because despite being at least ostensibly “gifted” my entire academic career, almost all of my professional success has been because I have been willing to climb steep learning curves at the expense of hours of my life and extreme frustration.
In short, I can relate because I too practice a type of “up close” magic which few people can even appreciate.
A good specific example of this that isn't widely known is the "muscle pass".
It amounts to holding a coin in a classic palm position (dead center of the palm), then spending several months strengthening the palm muscles and developing a callus, until you can propel the coin a foot or so without noticeably moving your hand.
> so much of it boils down to “I practiced for years and developed a seemingly superhuman ability to manipulate this object”.
Amateur magician here (but a magician member of the Magic Castle, so I know a few things). Manual dexterity is certainly a useful skill, but it's just a tiny part of the art of magic. You can be a consummate sleight-of-hand artist and still be a mediocre magician, and conversely, some of the best magicians don't use particularly difficult sleights. The best example of that I know of is Dani Da Ortiz's routine on Penn and Teller Fool Us:
Now, Dani is in fact a master at sleight-of-hand, and so one would think that you are watching a masterclass in sleights, but no. There are a few sleights, but they are not particularly difficult, beginner-level. I could do them, and I'm not particularly skilled. The trick is based almost entirely on timing and misdirection and psychological subtleties.
The degree of psychological subtlety in top-level magic continually blows my mind. The best magicians make it look like magic even when you know how the trick is done. Dani's trick is a consummate example. The performance looks like chaos, but in fact every detail is meticulously crafted and serves a purpose. It's almost like watching a dance.
" The best magicians make it look like magic even when you know how the trick is done. "
Agree with this. I remember being the 'skeptic kid' at a birthday party long ago. The magician involved me in a trick everyone could figure out but it was done with charm and I was completely delighted.
When teaching this whole routine, Dani spends much more time on the psychological background and nudges that are used, rather than the physical card manipulation techniques.
Yep. I am blown away not just by Dani's execution of this routine, but also by its design. It is a true thing of beauty, with layer upon layer of subtlety that most audience members will never appreciate because they don't know it's there. It is deliberately hidden. It has to be. Being hidden is an essential part of its function. Which is why I always try to seize the opportunity to raise awareness of this sort of thing among muggles.
BTW, if you are a Castle member, Carl Hein is currently doing a routine in the Library Bar that is IMHO in the same league Dani's Fool Us routine. We took some friends of ours to the Castle a week ago and Karl absolutely melted their brains. They're still talking about it. :-)
(And if you're not a Castle member but are in the LA area, contact me privately and I can set you up with a guest pass.)
That reminds me of a few episodes of Fool Us where they gave the prize to the guest because they knew exactly how the trick was done, but were literally unable to physically detect the person doing it because their skills were just that good.
A learning curve is a model for how much you can learn over time. A subject with a steep learning curve is actually one that doesn’t take much time to reach a high level. It’s the subjects with shallow slopes that take more work to raise up to proficiency.
It's a bit of a fossilized error. Like "this begs the question of...", "head over heels", "sleeping like a baby", and "Have your cake and eat it too" everyone should understand what it means, even if it doesn't stand up to technical analysis.
True, that meaning seems to be typical now. The opposite is the original meaning though.
> Scores of authors use the phrase “steep learning curve” or “sharp learning curve” in reference to a skill that is difficult to master. . . . Nevertheless, from the standpoint of learning theory, these and other authors have it backward, because a steep learning curve, i.e., a curve with a large positive slope, is associated with a skill that is acquired easily and rapidly (Hopper et al., 2007).
Everyone is talking about curves, but no one's talking about how the axes are labeled.
Based on common usage of the term learning curve, I had thought of it much like a power curve where the y-axis is the amount of cumulative effort you have to put in to reach a particular point on the x-axis, which measures mastery. Sounds like the official definition is effort on the x-axis and the total amount you've learned on the y-axis, which would indeed invert the meaning from how I've understood it.
Feels like the perfect thread to re-link to my favorite card-trick movie - Ricky Jay and His 52 Assistants. Not only is he a deft card manipulator, but he is a great story teller with a vast knowledge of magic history.
Ricky Jay was legendary, on the level of Houdini, and a gift to Hollywood as an adviser and occasional (but good) character actor. As good as he was at cards, he admitted others were better - but none beat him at throwing cards into targets. The man could probably have assassinated with a poison-laced deck, no lie.
He took on a young "partner" (mentored a student, but without admitting it), and presumably taught him much in his last years. Hopefully nearly all.
Jason is phenomenal. There is absolutely no telling how many hours he has had a deck of cards in his hands.
The foundations of card magic are certainly out there, and for me knowing how the fundamental slights work make it that much more magical to see it performed at such a level. No different than having played a sport makes spectating that much more interesting and appreciable.
fwiw, learning curve theory relates how much learning you get through time or episodes of experience. a lot of learning from a small amount of experience (easy to learn) means you will slide quickly down that steep learning curve. a shallow learning curve implies that it takes a great deal of experience to learn.
People downplay how important consistency is when trying to make changes to their lifestyle. Walking is good but whatever the exercise, a sustainable plan is best.
Same reason behind having a list when you go to the grocery store and sticking to it rather then buying whatever looks good then and there.
Had someone put up a project plan for something that was not disclosed as LLM assisted output.
While technically correct it came to the wrong conclusions about the best path forward and inevitably hamstrung the project.
I only discovered this later when attempting to fix the mess and having my own chat with an LLM and getting mysteriously similar responses.
The problem was that the assumptions made when asking the LLM were incorrect.
LLMs do not think independently and do not have the ability to challenge your assumptions or think laterally. (yet, possibly ever, one that does may be a different thing).
Unfortunately, this still makes them as good as or better than a very large portion of the population.
I get pissed off not because of the new technology or the use of the LLM, but the lack of understanding of the technology and the laziness with which many choose to deliver the results of these services.
I am more often mad at the person for not doing their job than I am at the use of a model, the model merely makes it easier to hide the lack of competence.
More seriously, you described a great example of one of the challenges we haven't addressed. LLM output masquerades as thoughtful work products and wastes people's time (or worse tanks a project, hurts people, etc).
Now my job reviewing work is even harder because bad work has fewer warning signs to pick up on. Ugh.
I hope that your workplace developed a policy around LLM use that addressed the incident described. Unfortunately I think most places probably just ignore stuff like this in the faux scramble to "not be left behind".
It's even worse than you suggest, for the following reason. The rare employee that cares enough to read through an entire report is more likely to encounter false information which they will take as fact (not knowing that LLM produced the report, or unaware that LLMs produce garbage). The lazy employees will be unaffected.
Strong agree. If you simply ask an LLM to challenge your thinking, spot weaknesses in your argument, or what else you might consider, it can do a great job.
This is literally my favorite way to use it. Here’s an idea, tell me why it’s wrong.
> do not have the ability to challenge your assumptions or think laterally.
Particularly on the challenging your assumptions part is where I think LLMs fail currently, though I won't pretend to know enough about how to even resolve that; but right now, I can put whatever nonsense I want into ChatGPT and it will happily go along telling me what a great idea that is. Even on the remote chance it does hint that I'm wrong, you can just prompt it into submission.
None of the for-profit AI companies are going to start letting their models tell users they're wrong out of fear of losing users (people generally don't like to be held accountable) but ironically I think it's critically important that LLMs start doing exactly that. But like you said, the LLM can't think so how can it determine what's incorrect or not, let alone if something is a bad idea or not.
Interesting problem space, for sure, but unleashing these tools to the masses with their current capabilities I think has done, and is going to continue to do more harm than good.
This is why once you are using to using them, you start asking them for there the plan goes wrong. They won't tell you off the bat, whuch can be frustrating, but they are really good at challenging your assumptions, if you ask them to do so.
They are good at telling you what else you should be asking, if you ask them to do so.
People don't use the tools effectively and then think that the tool can't be used effectively...
Which isn't true, you just have to know how the tool acts.
I'm no expert, but the most frequent recommendations I hear to address this are:
a) tell it that it's wrong and to give you the correct information.
b) use some magical incantation system prompt that will produce a more critical interlocutor.
The first requires knowing enough about the topic to know the chatbot is full of shit, which dramatically limits the utility of an information retrieval tool. The second assumes that the magical incantation correctly and completely does what you think it does, which is not even close to guaranteed. Both assume it even has the correct information and is capable of communicating it to you. While attempting to use various models to help modify code written in a less-popular language with a poorly-documented API, I learned how much time that can waste the hard way.
If your use case is trivial, or you're using it as a sounding board with a topic you're familiar with as you might with, say, a dunning-kruger-prone intern, then great. I haven't found a situation in which I find either of those use cases compelling.
reply