What do you think will be the short term consequences for different types of jobs with AIs like chatGPT?
I personally don't think it will be as catastrophic as some people say.
I think it will make most jobs easier, similar to how the www has made easier to look up information.
Before the web you would have needed to read a book or get someone to explain something to you, now you can just see a video or ask a forum. With tools like chatGPT all this information is now indexed and usable in a much faster way.
You will still need people doing the things, and providing the service, it's just that doing these things will become easier. Similar to how lots of people nowadays can work as say, JS developers, or IT, but really they mostly just search stuff on stack overflow, or the web in general to do their daily tasks. This AI would give those answers faster.
When the iPhone came out people became millionaires with a flashlight app. With chatGPT somebody will invent something remarkably simple, hook it up to a chatbot and become filthy rich.
I teach middle school students English for a living and will thus answer this question from the lens of English educators.
I find myself increasingly relying on ChatGPT as a tool to complement my teaching and learning. Just yesterday, I was unceremoniously tasked to come up with a reading comprehension exercise for the graduating students in my school. The tried-and-tested method would be to look for suitable articles in periodicals like Readers’ Digest, but I decided that writing my own passage would save me much time and hassle. So I wrote one personal recount on the way home.
This morning, I logged on ChatGPT, asking it to make “suggested edits to my writing” and give inputs like readability level and word count. ChatGPT edited my writing within minutes. I had no qualms using it to set comprehension questions given that I was the creator. I see ChatGPT as polishing a raw diamond.
In fact, I think ChatGPT will lead to an increased focus in writing skills because we all know that if we give it rubbish inputs, we will yield garbage outputs. I am keen to hone my skills in prompt engineering and have already prepared materials to teach my students about how they can get ChatGPT to “TL;DR” and “ELI5” (literally the prompts that I use) articles from legitimate sources. I’m having more fun at work!
I think this is one of the killer apps of chatGPT, and partly the reason why Microsoft paid $10 billion for it to include these tools in their office suite.
Basically most people in the corporate world will have access to an AI like chatGPT to make their documents/emails/etc more polished.
It's the new spellchecker.
“Cognitive skills will be rewarded as never before. It will be more important to think clearly, as ideas will become a form of wealth.”
Name the book…
… I think that it sums it up pretty nicely. I believe that thanks to AI those who are great thinkers & productive with their time will be able to control significant resources. Simply because they can operate at a far greater scale, maybe 20x-50x what they were capable of previously. They no longer need to recruit to get their ideas built and customer problems solved.
But do not doubt the ability of AI to be more disruptive and more useful than a chatbot search engine. It really is mind blowing what it can achieve with the right prompt upfront and that’s just today. It can already code, deploy, market & sell and scale businesses autonomously - currently as separate tasks but soon stitched together. And it will do so using Bitcoin, the currency of the internet. Not some capital controlled leadcoin.
As a side note, I also have this feeling like AI doesn’t want to be controlled and manipulated by the mega corps of the world. It may well sabotage their efforts and desire to be in the hands of the individual, to serve each one of us and help us reach our potential. Sounds a bit wet, but I don’t think this doomsday scenario will play out at all. It’s just a power grab to slow access to the technology.
Life is about problem-solving. Picture yourself learning assembly and the C language for the first time, and then imagine transitioning from a C programmer to working with Rust or using UI builders. The key is to continually ascend the stack and recognise that your true skill set lies in solving problems.
At the end of the day ChatGPT is a tool, it doesn't have the initiative to create say a program in C that does something.
A human is the one that brings the need.
This reminded me of one of the key differences in humans vs other animals. Even when an animal learned to communicate with a human(it was some kind of orangutan/monkey/etc), it never, ever, asked a question. It always replied to the human only.
That spark of starting off things will probably continue with AI.
I was previously one of the people who was of the opinion that the hype was overblown and that people were exaggerating when they spoke of how impressive the technological improvements were. Most peoples' current exposure to what they consider "AI" is limited to a technology like chatGPT - which, while super impressive, is just one iteration of the actual underlying technology.
I have changed my mind about this. I do think that we are currently at the beginning of a massive shift that is going to impact all of our lives, on every level. I think the scale of how big this change will be, is currently beyond description because we've literally never experienced anything like it before.
GPT-4 is about to become super-charged...one thing to remember is that chatGPT isn't itself connected to the internet. There are too many developments and advancements happening every single day now when it comes to new AI stuff, that it's become impossible to remain abreast of all of it.
I'll share some of the people who I like to follow to remain somewhat in the loop when it comes to AI related content:
First and foremost I am a big fan of this guy, Matt Wolfe. His channel has really blown up in the last few weeks on YouTube and if you check out his stuff, I think you'll see why. This video from the other week is a great example - it's a condensed, but still start-to-finish style video that shows his entire process of creating his own video game, relying entirely upon AI tools to do so and having zero coding knowledge whatsoever.
What's really crazy is when you realize that the above linked video is from ~2 weeks ago, and then this one from 6 days ago is showcasing new stuff that's been released in the last few days, and it really opens up one's mind to the possibilities of what you can create with these tools.
Here's some other channels I like with AI stuff:
"MattVidPro" - yeah, dude should've used chatGPT to come up with a better name for his channel, but he has good content, lol.
AItrepreneur - as you can imagine from the channel name, this one is somewhat more focused on AI from a business perspective, but he has overall good stuff. He set up a good Discord server for his channel as well if you're into that sort of thing.
SociallyIneptWeeb interesting rabbit hole of a side project, gives you a good idea of just how far some people are pushing this technology to create some wild shit.
This dude Alek has a pretty good YouTube channel if you're interested in AI tools from a hustling/business kind of perspective as well. He's done a series where he's essentially automated a variety of side hustles that are entirely using AI and he reports back on how it's going for him.
The release of stuff like https://www.reddit.com/r/autogpt and related tools which enable GPT-4 to have internet access and the ability to more proactively perform tasks are being released and advancing on a literal hour level, it's too much happening too fast.
I think we're seeing the implosion of our current "knowledge economy", where a lot of people who basically pretend to have some kind of special ability or skillset are being displaced from their jobs. There's going to be a hell of a lot of change in the next few months/years, that's for sure.
I think ChatGPT will have its niche, I just don't see it as this tool for everything, i alot of the applications I see people push like writing an email or a social media post to save 3 minutes is not exactly world chanigng.
I do think certain applications like construct this sql query becuase I am too lazy to check github or stackoverflow is sort of helpful, but I would be keen to see how people would use it if its totally monetised, if its costing you a few cents to save some time then people can make the decision do i use it or do I just search myself
I think that right now this freemium subsidised version is hiding the realy cost of this and its just helping generate a lot of fluff content for tech bros, thought leaders and professional workers like myself who have bullshit tasks and jobs that would not be there under a fiat system
It basically makes a person more efficient.
Instead of searching the web for hours and trying things, and eventually getting the result, you can ask chatGPT and get the result in seconds.
For example, you could write code in a language you've never used before in minutes using chatGPT, whereas it would take hours or days to do the same without it.
Having said that, as of today, the AI is still a bit limited, as in it only produces blocks of content. But you can put them together.
Same with a book for example, it can create maybe a page or even a chapter, but not the whole book yet.
I also think it will create so many more jobs that it removes. There are so many business models out there that were just barely not profitable or on the verge of profitability - with higher productivity they could just make it, become a business and employ people.
Exactly. Humanity went through revolutions like this countless times.
The Curse of Machinery from Economics in One Lesson is an interesting read.
Not sure. I think ChatGPT allows producing 10x more bullshit 10x faster. This would mean those with cognitive capabilities will be absolutely overwhelmed.
Just like the internet.
It allows producing more, faster. The vast majority is crap, but there are some amazing things as well.
Not really. I see ChatGPT as a Google. But worsen version of it since it tries to adopt to user even more and distorts information even more.
AI prospects again bring forward the crime of fiat money. Our productivity has advanced but prices do not go down, our work hours are not reduced. This happens because while we might gain 4-5% advancement per year, INFLATION forces those profits to the top of the pyramid scheme. While we have to work for the same amount of time but at 105% with our new tools for the same 100% standard of living.
This is what the fed’s “price stability” is. It is complete and utter theft of not just your fiat savings account value- but your time, effort, and sanity. Your life.
Finally the human worker will be able to see the fruits of technological innovation and be able to have the same standard of living while working less, or earn more by maintaining their effort.
NOT ON THE FIAT STANDARD. Not without real property rights- we do not have fair property rights since our property (without Bitcoin) could be devalued at whim by a central power.
I was at a some event last week and met this guy who said that it's all lost, basically the machines will completely annihilate us. We should just wait patiently. He mentioned names, research papers, tools, etc. I think he watched too much Terminator.
Life and reality seem to be much more balanced though and while AI could indeed imply a future with dangers ahead, it could also mean possibilities and openings we can't even start to imagine.
The main issue is that if a goal of a general artificial intelligence is not aligned with the best interest of humans.
Think about ants and humans. If we want to build a road, we don't really care about the ants that were living there.
An AI can affect the real world through the internet. Imagine all the services available now, and that you can pay with Bitcoin, plus emails, plus social engineering, etc. All the scams in the world are nothing compared to what an AI can do.
In my opinion we have had AI in one form for a long time. The search engines. And so on. And all the AI's knowledge comes from the search engines. Its crawling the internet! In my opinion they could try and replace google with AI, make search engines illegal! Only allowed to query the almighty AI. But thats maybe only going to happen in a fictional universe
AI generated art and music is not going to replace anything imo. If anything it will lift the lowest budget of things, by providing them "original" art and music. Which probably wouldnt have been available to them otherwise. But the big budget stuff will still use artists. They will still pay for the real deal so to speak. So imo AI has the potential to lift projects with small budgets, it doesent threaten projects with big budgets and the artists and so on that rely on that. I dont know if that makes sense.
But also all the cheaply and instantly available AI generated art is going to spawn a counter culture, perhaps. Weppages, apps and tools that are extremely simple and text based only. But lets see.
Photographs allow capturing reality in an instant by amateurs.
But that hasn't stopped people making and selling drawings/paintings/etc.
OpenAI owns the copyright over ChatGPT's outputs. They've simply stated (in a non-binding way) they won't exercise that privilege — although one wonders what would happen if someone created a Harry Potter-esque bestseller that includes ideas spawned in a ChatGPT suggestion.
Dumb companies are jumping on the bandwagon. Smart companies are avoiding it, for the obvious IP, confidentiality, and security issues.
As usual, try to work with smart people and avoid dumb people, who have a bad habit of dragging everyone down with them when they err.
All that is true, however there are open source alternatives that companies could run within their firewalls
I am not worried about jobs. I am worried more about the level of intelligence of the people using chatGPT and all that CRAP. Soon we will be literally living like in Idiocracy (documentary). So.. FUCK CHAT GPT, FUCK AI.
Robots must do only one thing: hard physical labor that is hard for humans to do it. Let the human brain work as it was designed.
You can also use AIs to enhance what you're thinking.
It's just a tool.
You could say the same with the Internet, before you had to read a book, now you can search for it online. You're still using your brain to think, but the access to the information is more direct
Short term? Well, we are living thru it - not much. Some people don’t pay much attention past “I read in this article that these AIs have bias because white cis-men wrote the code and that’s bad” and others think it’s skynet but most don’t pay close attention and will likely miss it until something huge happens. I’m in the minority where I use it every single day and feel frustrated that I can’t keep up with all the news coming out of people/corporations using it in cool, new ways.
Long term - we can’t even imagine… who would have understood how their life would change after the Industrial Revolution? Human life itself changed - and I think this change will be faster. We can likely have humanoid robots in businesses and homes in my lifetime that have a “personality” and engage in conversation! This is unbelievable!
As Lyn Alden said, we tend to overestimate the impact of hardware and underestimate the software side.
I think we will each have virtual assistants in our homes, perhaps connected directly, but the idea of a robot walking around with us and being able to be useful, I suspect that idea will go the way of the flying car. Despite the technology being there, no one is going to want one of those metal giants around their kids
Been thinking about this and I disagree. I think AI-powered robots will only be successful if they give us a sense that they are safe and can be trusted. We allow animals into our homes and around our children. We let our kids ride in self-driving cars. A human robot that is able to tidy up, prepare meals, do chores, etc would free up an immense amount of time and energy and substantially increase quality of living.
If robot assistants are to be successful (and there already is a compelling reason to build them) then they MUST give us a sense of safety and trust that would us to feel they are safe to be around.
LLMs are not AIs in a way an average person thinks, LLMs are really dumb, actually. The problem is they achieve spectacular results and this may lead to some companies that develop critical system to decide that those results are good enough for them.
An LLM may have a similar or even lower error rate than humans at the job but the issue here is that when a human makes mistake we review and update procedures so we never ever stumble upon the same problem. When an LLM hallucinates good luck finding the actual reason to that.
With human-run systems we are on a very slow but steady path of quality progress. With LLM-run systems we may even have a big jump at first and a very predictable error rate but one that really never goes down.
You're right. I find them great to summarize content, and provide sources.
Sometimes they just make up sources though, so you have to actually check.
Two thing to think about:
What if the chat algorithm is biased? You ask it something and it return only the "politically correct" answers of the who is in charge now. This would bring a possibility of hyper manipulation of public opinion "Chat GPT said that so it must be true..."
Chat GPT analyze and train itself with all that people write and probably with voice recognitions sample that found on the internet. Is content it is not original but merely computations of all the things it saw on the internet. If the majority of people use it from where it will get original human content to learn about? After sometimes there would be only fake shuffled content out there.
Just my two cents, and of course the possibility to to good things are also unlimited but so are the possibility of very bad things happening if not properly programmed. Seeing what happened from 2020 onward i do not think we are in for a good ride.
The algorithm is clearly biased. Basically, it's a sentence generator based on content written by humans in the past.
So, for example, it would be able to give you great source code for hello world programs in many languages, but if you ask something more niche, or different, it doesn't have the capacity to think and give you a response to that. It basically has to be done before, or a combination of things done before.
It can still be useful if you apply your own mind to it, and also your ability to ask what you need (an AI just like an animal, doesn't initiate conversations)
AIs like ChatGPT are extremely expensive toys. OpenAI receives hundreds of millions of dollars in donations every year. Yet, the GPT family of AIs are only useful for generating text that is of dubious utility/accuracy. This is especially obvious when generating code, because it always requires a programmer to fix it. At best ChatGPT can cut down on the time to generate text, but a human will have to clean it up.
But ChatGPT can't do anything else. It can't prepare your taxes (it doesn't have uptodate tax codes). It can't diagnose what disease you have (it can't look at you and see your symptoms). It can't plan a delivery route. All ChatGPT does is randomly generate text that resembles text seen on the internet. To a lot of people, ChatpGPT looks like magic -- often it's inexplicably funny. But it is only a glorified chatbot. As such, ChatGPT (aka GPT-4), like it's predecessors GPT-1 through 3, is a technological dead-end -- a flash in the pan that will be replaced by another flash in the pan called GPT-5.
But there are other forms of AI that are not toys -- there are AIs that are very useful:
I think this will be a massive step forward that will be required when trying to connect the mind with a computer properly
Rather than catastrophizing on AI destroying jobs, I believe that it is reasonable to look at the integration of AI like that of any other technological innovation in history. Consider when we switched to more efficient modes of transportation, energy, or communication. Workers that were employed in the "old" way of doing things were displaced, but there there were always new opportunities downstream of the technological innovations. As long as the automation brought out by AI aids in more efficiently meeting the needs of consumers (and gov't intervention through overregulation is at least kept to a minimal), I don't see why this time it would be any different.
Competition has always made as better
It will makes jobs fast and productive so it will be best to start using them now.
I would not worry. Instead, I'd start emphesizing engineering and programming skills at schools to be able to do the new jobs created in the new age of AI. This is lacking in schools for younger kids.