I'm a software developer who did AI research in undergrad and have kept up with papers involving AI, specifically LLMs and the performance metrics published by major AI players. I think it's important to remember that the people who currently are attempting to convince every industry leader that AI will not just replace people, but will be better in every way, stand to make a lot of money in the short term via investment. I think it's also important to contextualize the performance of these leading models. Without getting too specific, it should be known that when someone says "The AI is now PhD level" what they are actually saying is generally "When answering questions you would ask someone with a PhD, when the AI model has been trained on the questions AND their answers, the AI can reach as high as 70%-80% accuracy." Which is a lot less impressive sounding. Why would they not tell you those important details? Because it directly affects their bottom line. It's also important to know that AI does not "create" in the traditional sense. It remixes. This is especially notable with code. While it can code things that have known and well documented solutions fairly well, it is terrible at creating novel solutions to novel problems. This is not just because it's not "smart" enough, this is a limitation of the design. LLMs, which all major AI are based on, work by basically weighting outputs based on the inputs, the inputs being large amounts of data, mostly scraped from the web. Problems that don't come up often or that are unique will make the AI struggle. There is no known mechanism for overcoming this, as it is an issue of design. We would need to invent an entirely new method of creating AI to overcome this. I think AI is an incredible tool that will become a part of development in many areas; art, software, architecture, robotics, music, etc. But especially when we're talking about artistic endeavors where things like feel, atmosphere, artistic direction, uniqueness, etc are not just important, but the most important difference between success and failure, humans will be better at these tasks than AI for the foreseeable future, no matter how good LLMs become.
Very insightful response. It will be a long time before AI is able to replace senior level engineers, coders, designers, and artists, but I don't think it will be too long before it is good enough to do the work of entry level hires in those fields, which would pose a problem if company's actually capitalized on that and outsourced those positions to AI. It's already hard enough as it is to find a job in tech (and more specifically, game development) with little or no experience because most companies nowadays expect that going in, so if all the base level roles are essentially automated then fresh grads are going to have an even harder time getting the experience they need to move into more senior level roles.
@Ryan-gm6ri This is true. But I think companies already consider hiring new grads as a cost of doing business. No new grad is really worth what they're being paid, and even though they're often seen as code monkey, it can be 6 months before they're even familiar enough with a code base to actually be useful and trusted to write code, and even that code will be vetted by mid level engineers, etc. I think the future will probably be Jr. And Mid level engineers using AI to write code and rapidly prototype, while Senior level engineers do what they've always done: make high level decisions based on their expertise and jump in when there's critical issues or challenges that need an expert's opinion. Entry level jobs will always be rare in tech, but the good thing is that AI open source models mean that these new grads will have bigger and better projects to show off their higher level design skills, and demonstrate their potential so that these companies choose to invest in them. I think the days of job hopping from one place to another looking for a bigger paycheck will end, and instead companies will expect "loyalty" when they invest time in training you, and things like 3 and 5 year contracts will become the norm rather than Google having the same turnover as Walmart or McDonalds for first year developers like they do now.
use deepseek and huyuan 3d to begin... trust me guys holy sh*tt deepseek to help you understand codes and huyuan with the assets. only thing you have to know is knowing how to use unity and how its machine works
I hope its used as a tool to help future game devs reach their imaginative goals more efficiently. The main tip we hear from every published indie game dev is to start small. Basically make something you probably didnt want to make, just to have a product out there. AI can potentially make your goals of a game easier. I dont think itll be at a point any time soon where it can be imaginative and replicate the human touch in a games development.
I think it all depends on how you think AI will be used. Conversations about AI seen to go to a metaphysical level really quickly as the abstract idea of artificial intelligence is easier to deal with in this metaphysical, abstracted world of 'what is intelligence and creativity anyway?' While in reality, the practical applications of the tool of AI have much more profound implications that will affect us all in the industry. We see that with image generative AI there has been an exponential growth of AI generated digital media. While the discussions about whether it's actually creative or just stealing are interesting, they don't negate or preempt the fact that marketplaces offering digital images have been inundated with AI generated content - everything from UE/unity marketplace to Adobe stock images has turned into a digital dumping ground. This digital dump leads to a flooded marketplace, which affects everyone. The same will be true of game development, where genuinely crafted gems will have to compete with AI generated slop for digital shelf space. It's not that it will take your job 1:1 and personally replace you. It's that the return on investment in you and in your work will diminish to the point where even customers or consumers won't value your contributions.
i think it's risky to use ai to make a game, there's a lot of big industry games completely failing from taking shortcuts in the last 2 yrs , if they all start turning to ai i feel it will just get worse. good for them tho, i say let the corporate game world die off. i truly will not miss it. there are still going to be good companies out there making killer games like From Soft and Rockstart but a lot of companies making bad decisions now. Indy Devs rise up! this is your time
I had a funny interaction with Grok trying to get it to generate a cyberpunk fire engine. It tried, but wasn't getting it. I told that is was missing the mark. After several attempts I told it they still weren't correct, actually they were getting worse. It replied that it is just a large collection of lines of computer code and doesn't have human intuition, it is code, no human. That was good for a laugh.
No, AI will help any random noob become an indie game developer. But AI will destroy human creativity (it's already doing it) by doing creative things on its own. There will probably be some old-school creators left here and there, but for the most part, people will just be lazy consumers of AI products.
I used the help of AI to learn how to code my game but I did not use the AI to make any creative decisions or actually implement anything straight in the game. So I do believe there is a place for it and it helped me accomplish my dream, but that's not saying people cant use it the wrong way. As always great video!
I feel like AI should be regulated to the point where we are given a stamp or some type of label saying AI was used x amount of percent so we can choose to support or buy said products etc
No, its a tool. It will probably end up destroying a few studios due to misuse and trying to use it to replace the actual devs and artists but when people figure it out properly it will help with productivity massively as well as expanding worlds and making them more immersive.
@@SpaceRanger37 honestly i find its used best as more of a search engine/knowledge base currently. The current AI we have isn't really AI anyway its more like a probability algorithm so while it technically meets the definition we have if people understood it wouldn't really count as AI
Exactly. Treat it as a tool to assist with your work and never overuse it because if you use AI for everything is it truly your work at the end of the day.
If you're afraid of AI replacing you, you're a bad dev. It's like a construction worker being afraid of drills and hammers. A competent builder sees them for what they are: tools, they don't go through an existential crisis when thinking about them
He is largely talking about current generation AI. We are still extremely early. Current AI can already disrupt a lot of jobs in gamedev. As he mentioned 2d concept artists are having trouble finding contract work already. Tools are getting very good at writing code. If you're not trying AI to write code for your game, try it. Music is getting good, 3D is just getting started. Over just the next few years AI will get very good at everything, and if you're worried about something as complicated and creatively driven as gamedev, think about the jobs most people have and how easy AI and robots will be able to replace them. The big question is not how long until AI replaces game developers, it is how long before it replaces enough jobs to raise unemployment to unsustainable levels and we have to enact something like UBI to keep the economy from collapsing.
Indeed! AI is evolving so quickly, it's hard to keep track. Not too long ago, no one thought AI could be truly creative, yet now it’s creating art, music, and poems. The advancements are incredible, and with the efficiency and cost improvements, the possibilities are massive. But I still think there will be areas where humans have the upper hand for a while, like in creative direction or tasks that require emotional nuance. Fields like UI/UX design or overseeing a project’s vision might stay more human-focused for the time being. What do you think? Which areas should indie developers focus on to stay relevant as AI keeps progressing? Or do you think UBI will happen faster than we expect?
If we hit unemployment of 35%+ we're looking at socio-economic collapse. It's a complete fantasy is that everyone is going to be collecting UBI, living in big houses in nice neighborhoods, enjoying life. That's not going to happen.
@@fgama I think you have a fundamental misunderstanding of what AI is, and what 'creativity' is... AI is not 'evolving.' Living things evolve. AI is not alive. It has no concept of meaning. It's an algorithm, which finds patterns in data, and generates output based on statistical probability. In order to be creative, one must make a series of creative decisions. AI cannot, by nature, be creative. What you perceive as AI's intelligence and creativity is actually human creativity and intelligence - which AI has trained on. It's human beings behind the data who are creative. AI is a mimic, nothing more.
@ We are headed for 100% unemployment. In coming years AI and robots will do every economically valuable task better, faster, cheaper, and safer than humans can. The question is how we manage the transition to that point as unemployment rises due to increased automation. UBI is the only solution I've heard so far.
@@benpielstick I think you have an overinflated sense of transformer-based AI and what it is/will be capable of, but I appreciate your opinion. Personally, I think we've seen the limit of what that architecture is capable of. We'll see how it all shakes out.
Seems to me like everyone who currently works in digital asset creation (software, design, writing...) is currently trying hard to convince themselves that they will not be replaced by AI. The truth is that two years ago these things were little more than predictive text engines but now we have multi modal, chain of thought, tree of thought, huge context windows, larger models, agentic behavior, and now the efficiency (training and inference) and therefor cost is dramatically improving. The only thing currently holding these models back is high level architecture and design capability, but that's coming soon, and once it arrives every gamer or creative individual will also be a potential game developer.
Yeah, I get what you mean. AI’s moving so fast, and it’s almost hard to keep up. A few years ago, people didn’t think AI could be creative, but now it's generating art, poetry, and even music. It's crazy how much it's advanced, and with the improvements in efficiency and cost, the potential is huge. But I still think there are areas where humans will have an edge for a while, like unique creative direction and areas that require more emotional depth. Maybe things like UI/UX design or the overall vision of a project are safe for now. What do you think? What areas should indie devs focus on to stay ahead as AI keeps evolving?
@@fgama To be clear, I don't think AI will replace humans, there will just be many more humans who are capable of the same things since having deep technical knowledge or years of experience will no longer offer you much of a competitive advantage. Short term I can imagine the market becoming very saturated and the people who succeed will either have the best marketing strategies or simply the best most creative ideas. In the long term it's hard to make predictions, this tech is going to reshape entire economies... we all want to protect ourselves, but it's much bigger than our individual concerns.
@@invalid_unique_handle Loved this! So it will be more like "know how to cut through the noise and stand out in a sea of options". But thinking about it, even if AI gets as good as a human and all things automated, humans will still be able to compete with it in the market, specially in indie game dev. But more competition indeed. But let's hope AI will bring way more advancements and good things for our civilization than our individual concerns are focused on indeed.
If you lay in bed. And we replace every molecule in your body one by one, is you still you? Now instead of replacing the molecules in place, lets shift them half the width of a molecule to he left. Are you you? How about shifting everything one full width of a molecule. Are you you? When do you stop being you? Can we shift your body copy a meter to the left? I don’t know the answer, but the result would probably answer the questions if we could teleport someone by copying them. What if we don’t delete the original?
Assuming that our subjective experience is the direct result of our neurons firing and that there isn’t some metaphysical “soul”, replacing the original you with an exact copy wouldn’t change anything. It helps to think of subjective experience as a process that only exists in the present, rather than an individual“thing”. By this logic, creating an exact copy without deleting the original would result in a 50% chance of your consciousness seemingly teleporting into your clone!
I use AI to help me overcome obstacles in achieving my goals. It's definitely useful for that. However, I found that combining RUclips videos and refining AI outputs tailored to my needs works best. That said, I don't use AI for art. For my game, Bubble Tank Frenzy, I used AI only to generate concept art for my tank model. I then shared that concept art with a Fiverr 3D artist who brought the tank model to life. As for pixel art, well, let's just say I avoid using AI for that.
No because I avoid lazy people and ai anything created or has been assisted by ai I try to avoid simply because technology and ai is making people lazy and pushing out sub par products
What GPT knows so far is the initialization of the creative process, every idea GPT comes up with is a "first thought", rather than an abstract one. These training models have a hard time with 'Quality Abstraction' but are now hungry for learning the logic behind Quality Abstraction to create innovative ideas. When it figures it out, I'm afraid the more intelligent versions of GPT will be pulled like a rug. Best thing to stop AI in it's tracks is EXCLUSIVELY use it for research purposes.
I think AI tools will aid in the creation of junk food while hand made things will start to have more value. I understand the industry is moving towards the use of tools to speed up the process but this will move to a fast food approach to mass development as they arent using AI to make games like Balatro or Lethal Company, but to make the games they currently are making but faster. They might be able to risk on new ideas but they just want a fast way to make huge block busters.
Well, for the people making shavelware and asset filps absolutely, but you dear viewer reading this would never be such a bottom feeder like them would you. so you're completely safe and can now sleep easy.
Kevin doesn't understand that hardware is not the software...we are not expressions of our molecules. The software is not the consciusness...the cosciusness generate the software called "mind" and this software uses the hardware called body. To experiment this you need to be able to do deep meditation and lucid dreaming...than you can understand the reality. AI has no consciusness and will never has it...but the problem is that it can really substitute a lot of people in the work. Another problem is that we will have a lot of garbage made with ai...also in indie game sector.
No personally I don't think AI could ever replace real talent even though AAA companies will try cause all they care about is money not the quality of the games they create
AI is an algorithm which recreates learned patterns based on input tokens. It's not alive; it's not intelligent. It doesn't understand concepts. So no, AI isn't going to destroy Indie game dev. The real risk comes from the fact that it will be used to generate so much low-effort shovelware that it'll be even harder to get your game noticed - but that's a broader problem facing the internet in general.
I agree. I don't want to get complacent about AI replacement since a large portion of the population genuinely don't seem to have standards for quality, but I think fundamentally people seek out things made by people. I think AI can be a good tool to help solo devs get their start, and the shovelware problem can be mitigated somewhat by making curated sites that weed out games made with AI.
@@Nightmartlet Man, to this day I firmly believe that most games on Google Play are just used for money laundering, because I am unwilling to admit that human beings are so low intellect that they eat up all that slop AND then rate it 4.9 stars.
You can tell AI to make a picture in the style of Claude Lorrain, and it will do a pretty good job probably. But it wont have the soul, the intricate storytelling, or the feel of the real paintings. In the same way, if or when AI can get to the point of making entire games by themselves, keeping the continuity of characters and story intact, having a cohesive art style, and a user experience that doesn't want to make you turn it off - it will still lack soul. The best art is more than a picture, more than words on a page. For a art as a game its more than the polygons, the logic and story. The best art has something that humans connect with at a core level. I don't think AI will get to the point of really being able to connect to a human with art. I'll start worrying when AI can - completely by iteself - make a game as full of soul and story as Witcher 3. Until then, its a useful tool for helping to speed up tedious coding tasks or for assistence debugging, or for perhaps some ideation for an art style. I don't think soul is in our chemicals or DNA. I think we receive inspiration from something else that science is yet to fully acknowledge or understand. I think Rick Rubin has a pretty good grasp on what it is and how to use it.
Seriously,there were ton of things that had destroyed many things else,it really depends.We should accept the art forms from the A.I softwares,they are sometime good and useful.
Ok, but imagine I don’t disintegrate you, and I print a duplicate of you just next to you, and then I disintegrate you. Is the duplicated you you? No. But what’s the difference between them and you? None.
This. We've got way better tools & resources available than at any time in gaming history, so more people can join in. But most games - indie & AAA - are still just not very good 😅 GenAI is good at giving you a kinda-sorta okay version of what you asked it for, but it can't give you what you don't know to ask for. If you lack the skills & ingenuity to create a good game, it can't help you.
It will just mean more entrants, everyone will be more efficient. So I guess more competition since these tools are great--e.g., I have used chatgpt to help me build certain mechanics in godot. Eventually there will be processes that can build MOST of a game I guess, but without tweaking and guidance it will never be good art. So I guess on the bright side: it's easier to bring ideas to life now. Maybe the secret sauce will be how you make yourself unique, really drill down into the "hook" concept and work on your launches?
Really enjoied your free video with tips and i totally going to buy your game course in few months cause life hitted me like hurricane lmfao (really want to know how to make games and learn it and be consistent on it) i use godot cause its free and i really enjoied that u have an roadmap for 1 year to make games and really enjoied your story so thank you for being u man thats it. like if i could make games it will be f dope and if they are good games they are f dope i have some ideas in my head more like an idea but dont know how to make it into reality its like everything is black and i am searching lmfao with my hand for something (cause i dont really know shi@t honestly) but ya
In the 1900 people were scared and annoyed that some technology will replace them, robots and machines has already replaced people in the past, without apologies. AI will definitely do the same thing. Get good with AI everyone. This is the only safety net here. History is bound to repeat it's self, so don't try to fight it.
The way I see ai is like another person helping me, giving me tips and for generative ai, I’d use it for placeholders, like music, models and anything else it can make. For a final product I’d rather make everything myself tho
We have game engines now that allow one person to make a game, which in the past would take tens/hundreds of people. You don't see anyone crying about jobs lost due to game engines.
Indeed! Now AI is doing the same with jobs in creative fields, including game dev. It is pushing the envelope in ways we never thought possible. It’s not just about creating games, it’s about adapting and finding where humans still add that special touch.
For now, I think areas like creative direction, emotional nuance, and overseeing a project’s vision will likely remain more human-centric. What do you think? Any ideas of what should indie devs focus on to stay relevant as AI advances?
@@fgama I see AI as a productivity tool, just like game engines. If used correctly it will allow us to make better games with less effort. The advent of game engines didn't result in some destruction of gaming jobs and neither will AI.
@@vdimension6300 Hmmm I'm not sure about that. I think AI is more agressive than game engines and, let's say, has more "automation power". And in the past there was less competition in the game industry too. As you said, to " make better games with less effort" will increase competition too. But I liked your view on it! Let's hope it will bring more opportunities and who knows even entire new markets we hadn't even thought yet!
@@fgama What I am trying to say is that when game engines were introduced, suddenly so many potential jobs for developers were lost. But no one complained about it because these same developers could then become gameplay developers for example. So you lose one job but other jobs are created. Same logic goes for AI. You may need a lot less concept artists for example, but these same artists can reskill to become artists for the actual final art used in the game. And if AI can do that too, which I doubt it could anytime soon, then they can always re-skill for something else. Most people used to work on farms, and suddenly farming machines meant 1 person could do the work of 100 people. Does that mean no one has a job nowadays? And if AI somehow reaches a point where it can do anything a human can do, then we're living in a utopia where everyone lives like a millionaire with free stuff produced by AI and everyone is free to pursue their passions. Like making games without the help of AI ;)
He has a materialistic worldview but doesn't know where creativity comes from? That seems like a huge gap in his beliefs. How does he think we have free will either?
Will AI Destroy Indie Game Development? No, on the contrary, it will definitely make it better. AI can be a great help in many scenarios, but it will never replace the human factor, especially when a game aims to deliver a unique experience. Many players don’t choose a game just for its visuals or gameplay; there is always something that truly hooks them. That emotional connection is something AI cannot deliver. However, as a tool, AI can greatly assist creators by broadening their vision and optimizing processes. It’s like building a house: you can do it with basic tools, but having better tools makes the process easier. Still, what truly defines the house is not the tools but the needs and emotions of the person who will live in it. The same applies to video games-AI can help, but without human consciousness and creativity, it cannot capture the essence that makes a game truly connect with players.
Ohh the have a 100 monkeys write shakespear. Good question where does the creative spark comes from and do we connect to that spark in a way to create and enjoy creativity
Hello TB, been following u for a while, I have a video request, can u kindly make a video about game character design, that is designing my character( tips)
What's going to happen to gaming is what's happened to music. Once game development becomes easy and cheap there will be 100 million game developers flooding the market with autotuned games. Then there will be a culling where there's a few huge game companies the Taylor Swift of gaming and all the rest will be fighting for views.
How relevant with the Deepseek announcement. Do your opinions change after that announcement of a cheap, Chinese knockoff AI? If they can do this cheaply, what is to stop that company from going further with their development? Do you get what you pay for instead?
The man who wrote the worst bioshock sequel and its dlc is one you shouldn't take game dev advice from. Infinite's story was nonsensical and the dlc retconned too much of it.
@@HOM-01 Nah, this video was mostly discussing if AI can replicate existing works and a discussion of the "Teletransportation" paradox which is a high school philosophy thought experiment. Doesn't really speak to anything about the impact of AI on the industry.
I don’t know, man… I partly agree, but idk... AI advancements are just so rapid, they’re almost exponential. A few years ago, no one believed AI could be creative, producing art, poems, and even music. But now it’s advancing so fast, with unexpected booms of innovation seemingly coming out of nowhere. Even though it still can’t truly understand the creative process or invent something completely original, AI can now replicate Shakespeare’s style of writing really well! I think AI will reduce the need for as many developers to code, at least. To be safe, maybe we should focus on areas like UI/UX, things AI might not dominate as much. 🤔 So what skills in indie game development do you think are the least likely to be replaced by AI? How should creators future-proof themselves to stay relevant as AI evolves? Any ideas on which secure areas we should invest time in?
IDK about the coding part, my experience has been mixed. The least likely positions are something like Tech Artist and actual developers, if you're on the art side of things well . . . But personally I don't buy the hype of AI, it's just another push from investors to rake in tons of cash on something that we'll find out isn't all that profitable, as is the cycle in Silicon valley. You talk about the growth of AI, yes it has come far but from what I see and know, it'll slow drastically, some say it already has. I suggest looking up other opinions. There are several videos on youtube that cover this topic.
@@D.KRyley-mq1do Thanks for you input! It makes sense indeed. I get there is a lot of hype and fake propaganda, but I still think AI is quite unpredictable, it could still have a boom out of nowhere. As you agreed the art side is already being replaced... It could take a while, but I just like to be safe because it can be so sudden. So just for the sake of paranoia lol, I'd love to hear which specific parts of a tech artist and developer do you think would be the safest?
@ Tech artist are needed to create something new, a new innovative look comes, typically from tech artist. A step up from them are graphic programmers: Another field that cannot be replaced by AI anytime soon because again it requires creative and NEW innovative approaches to make the GPU do GPU magic. Developers are the same way: the role requires creative thinking in order to meet certain objectives. Imagine a world where developers never did anything new, just used what worked before. And this is the issue with AI, is that IF it ever gets to the point where it can be completely creative and becomes the AI that people think of when they think AI. Then everyone's job is on the chopping block, like yesterday. But that isn't the case and when and if that does happen society has a lot bigger issues to deal with and everyone will be in the same boat. To me fearing job replacement with AI is like fearing a US economic crash. Possible, sure but if/when it does happen everyone's hurting. My question to you is, if AI can do all the things you think it will be able to do in the future. Then whose job is actually safe? If the answer is none, they why worry about it? Just do you.
@@D.KRyley-mq1do The art is terrible -to- mixed quality as well. Like with coding, you need to have a deep understanding of what you're trying to create, and specify & iterate to such a degree that you need to be a competent artist - someone who would already be faster without it. Also like coding, unskilled people are very bad at describing exactly what they want, picturing something from nothing, and getting the model - efficiently! - to create pictures from their mind's eye. There's also the big issues of lack of consistency, inability to stay within project constraints, and legal uncertainty (genAI training data is "polluted" with info you don't have rights to, and can produce things that open you to liability). You also can't sue the model or its developers if it does something that hurts your business. From a purely commercial standpoint, it's just not as helpful or cost-effective as simply hiring people (artists are cheap, and coders outside the US are as well).
AI will destroy everything actually, but you will be able to make a AAA game alone without knowing programming art or music, as long as you can read and write you will be able to make a game
LLMs are a word guessing machine. AI is a guessing machine, that's it. It cannot and never will be, able to think for itself. It's not sentient. You ask it to do something, it spits out a known way to do it. It's a tool for people to use to shortcut the learning process of coding and bring their vision to life But the true S tier games still come from people that have both vision and coding skill.
It is the usual copium that people keep telling themselves. There is no law in nature that says only flesh and blood beings could be sentient, or only sentient things may generate novel things.
@@N7sensei the failure of neuroscience and psychology to properly define consciousness displays how behind/egotistical modern science is with the mind. I'd rather trust a hyper intelligent ape with controls to create something novel in some art form than a billion chatgpts combined. Do you know how LLMs work? It's a large language model. It's called machine learning. The machines learn by looking at data that they are trained on. Without the data, they are nothing. YES, they can REMIX the data to create "NEW" games in novel ways but that's just a REALLY shiny remix. Enjoy your stupid remixes I guess.
I'm a software developer who did AI research in undergrad and have kept up with papers involving AI, specifically LLMs and the performance metrics published by major AI players.
I think it's important to remember that the people who currently are attempting to convince every industry leader that AI will not just replace people, but will be better in every way, stand to make a lot of money in the short term via investment.
I think it's also important to contextualize the performance of these leading models. Without getting too specific, it should be known that when someone says "The AI is now PhD level" what they are actually saying is generally "When answering questions you would ask someone with a PhD, when the AI model has been trained on the questions AND their answers, the AI can reach as high as 70%-80% accuracy." Which is a lot less impressive sounding. Why would they not tell you those important details? Because it directly affects their bottom line.
It's also important to know that AI does not "create" in the traditional sense. It remixes. This is especially notable with code. While it can code things that have known and well documented solutions fairly well, it is terrible at creating novel solutions to novel problems. This is not just because it's not "smart" enough, this is a limitation of the design. LLMs, which all major AI are based on, work by basically weighting outputs based on the inputs, the inputs being large amounts of data, mostly scraped from the web. Problems that don't come up often or that are unique will make the AI struggle. There is no known mechanism for overcoming this, as it is an issue of design. We would need to invent an entirely new method of creating AI to overcome this.
I think AI is an incredible tool that will become a part of development in many areas; art, software, architecture, robotics, music, etc. But especially when we're talking about artistic endeavors where things like feel, atmosphere, artistic direction, uniqueness, etc are not just important, but the most important difference between success and failure, humans will be better at these tasks than AI for the foreseeable future, no matter how good LLMs become.
Very insightful response. It will be a long time before AI is able to replace senior level engineers, coders, designers, and artists, but I don't think it will be too long before it is good enough to do the work of entry level hires in those fields, which would pose a problem if company's actually capitalized on that and outsourced those positions to AI.
It's already hard enough as it is to find a job in tech (and more specifically, game development) with little or no experience because most companies nowadays expect that going in, so if all the base level roles are essentially automated then fresh grads are going to have an even harder time getting the experience they need to move into more senior level roles.
@Ryan-gm6ri This is true. But I think companies already consider hiring new grads as a cost of doing business. No new grad is really worth what they're being paid, and even though they're often seen as code monkey, it can be 6 months before they're even familiar enough with a code base to actually be useful and trusted to write code, and even that code will be vetted by mid level engineers, etc.
I think the future will probably be Jr. And Mid level engineers using AI to write code and rapidly prototype, while Senior level engineers do what they've always done: make high level decisions based on their expertise and jump in when there's critical issues or challenges that need an expert's opinion.
Entry level jobs will always be rare in tech, but the good thing is that AI open source models mean that these new grads will have bigger and better projects to show off their higher level design skills, and demonstrate their potential so that these companies choose to invest in them.
I think the days of job hopping from one place to another looking for a bigger paycheck will end, and instead companies will expect "loyalty" when they invest time in training you, and things like 3 and 5 year contracts will become the norm rather than Google having the same turnover as Walmart or McDonalds for first year developers like they do now.
"AI won't take your job. A person using AI will take your job". ------Genghis Khan
Did Genghis Khan really say that?
@@TheSophieGames Yes.
Sure it wasn't Genghis Tron?
use deepseek and huyuan 3d to begin... trust me guys holy sh*tt deepseek to help you understand codes and huyuan with the assets. only thing you have to know is knowing how to use unity and how its machine works
Hands down, one of the top 5 Khans in my book. Smart lad.
I hope its used as a tool to help future game devs reach their imaginative goals more efficiently. The main tip we hear from every published indie game dev is to start small. Basically make something you probably didnt want to make, just to have a product out there. AI can potentially make your goals of a game easier. I dont think itll be at a point any time soon where it can be imaginative and replicate the human touch in a games development.
I think it all depends on how you think AI will be used. Conversations about AI seen to go to a metaphysical level really quickly as the abstract idea of artificial intelligence is easier to deal with in this metaphysical, abstracted world of 'what is intelligence and creativity anyway?' While in reality, the practical applications of the tool of AI have much more profound implications that will affect us all in the industry.
We see that with image generative AI there has been an exponential growth of AI generated digital media. While the discussions about whether it's actually creative or just stealing are interesting, they don't negate or preempt the fact that marketplaces offering digital images have been inundated with AI generated content - everything from UE/unity marketplace to Adobe stock images has turned into a digital dumping ground.
This digital dump leads to a flooded marketplace, which affects everyone. The same will be true of game development, where genuinely crafted gems will have to compete with AI generated slop for digital shelf space.
It's not that it will take your job 1:1 and personally replace you. It's that the return on investment in you and in your work will diminish to the point where even customers or consumers won't value your contributions.
AI will just spam the s*** out of the internet with low quality garbage.
Treat it as a tool to assist with your work and never overuse it because if you use AI for everything is it truly your work at the end of the day.
i think it's risky to use ai to make a game, there's a lot of big industry games completely failing from taking shortcuts in the last 2 yrs , if they all start turning to ai i feel it will just get worse. good for them tho, i say let the corporate game world die off. i truly will not miss it. there are still going to be good companies out there making killer games like From Soft and Rockstart but a lot of companies making bad decisions now. Indy Devs rise up! this is your time
I had a funny interaction with Grok trying to get it to generate a cyberpunk fire engine. It tried, but wasn't getting it. I told that is was missing the mark. After several attempts I told it they still weren't correct, actually they were getting worse. It replied that it is just a large collection of lines of computer code and doesn't have human intuition, it is code, no human. That was good for a laugh.
No, AI will help any random noob become an indie game developer. But AI will destroy human creativity (it's already doing it) by doing creative things on its own. There will probably be some old-school creators left here and there, but for the most part, people will just be lazy consumers of AI products.
I used the help of AI to learn how to code my game but I did not use the AI to make any creative decisions or actually implement anything straight in the game. So I do believe there is a place for it and it helped me accomplish my dream, but that's not saying people cant use it the wrong way.
As always great video!
I feel like AI should be regulated to the point where we are given a stamp or some type of label saying AI was used x amount of percent so we can choose to support or buy said products etc
I probably won't be buying too many games that are made completely by A.I. I'd rather support human creations. Machines don't need the money.
Only if it's uncensored
No, its a tool. It will probably end up destroying a few studios due to misuse and trying to use it to replace the actual devs and artists but when people figure it out properly it will help with productivity massively as well as expanding worlds and making them more immersive.
Exactly my thoughts, I imagine that creators will be using ai to help with coding and finding bugs
@@SpaceRanger37 honestly i find its used best as more of a search engine/knowledge base currently. The current AI we have isn't really AI anyway its more like a probability algorithm so while it technically meets the definition we have if people understood it wouldn't really count as AI
Exactly. Treat it as a tool to assist with your work and never overuse it because if you use AI for everything is it truly your work at the end of the day.
yeah A.I will improve the performance at some aspect but it destroys the natural competitive advantages.
If you're afraid of AI replacing you, you're a bad dev. It's like a construction worker being afraid of drills and hammers. A competent builder sees them for what they are: tools, they don't go through an existential crisis when thinking about them
Ai will definitely replace devs in the future
He is largely talking about current generation AI. We are still extremely early. Current AI can already disrupt a lot of jobs in gamedev. As he mentioned 2d concept artists are having trouble finding contract work already. Tools are getting very good at writing code. If you're not trying AI to write code for your game, try it. Music is getting good, 3D is just getting started. Over just the next few years AI will get very good at everything, and if you're worried about something as complicated and creatively driven as gamedev, think about the jobs most people have and how easy AI and robots will be able to replace them. The big question is not how long until AI replaces game developers, it is how long before it replaces enough jobs to raise unemployment to unsustainable levels and we have to enact something like UBI to keep the economy from collapsing.
Indeed! AI is evolving so quickly, it's hard to keep track. Not too long ago, no one thought AI could be truly creative, yet now it’s creating art, music, and poems. The advancements are incredible, and with the efficiency and cost improvements, the possibilities are massive.
But I still think there will be areas where humans have the upper hand for a while, like in creative direction or tasks that require emotional nuance. Fields like UI/UX design or overseeing a project’s vision might stay more human-focused for the time being. What do you think? Which areas should indie developers focus on to stay relevant as AI keeps progressing? Or do you think UBI will happen faster than we expect?
If we hit unemployment of 35%+ we're looking at socio-economic collapse. It's a complete fantasy is that everyone is going to be collecting UBI, living in big houses in nice neighborhoods, enjoying life. That's not going to happen.
@@fgama I think you have a fundamental misunderstanding of what AI is, and what 'creativity' is... AI is not 'evolving.' Living things evolve. AI is not alive. It has no concept of meaning. It's an algorithm, which finds patterns in data, and generates output based on statistical probability. In order to be creative, one must make a series of creative decisions. AI cannot, by nature, be creative. What you perceive as AI's intelligence and creativity is actually human creativity and intelligence - which AI has trained on. It's human beings behind the data who are creative. AI is a mimic, nothing more.
@ We are headed for 100% unemployment. In coming years AI and robots will do every economically valuable task better, faster, cheaper, and safer than humans can. The question is how we manage the transition to that point as unemployment rises due to increased automation. UBI is the only solution I've heard so far.
@@benpielstick I think you have an overinflated sense of transformer-based AI and what it is/will be capable of, but I appreciate your opinion. Personally, I think we've seen the limit of what that architecture is capable of. We'll see how it all shakes out.
Seems to me like everyone who currently works in digital asset creation (software, design, writing...) is currently trying hard to convince themselves that they will not be replaced by AI. The truth is that two years ago these things were little more than predictive text engines but now we have multi modal, chain of thought, tree of thought, huge context windows, larger models, agentic behavior, and now the efficiency (training and inference) and therefor cost is dramatically improving. The only thing currently holding these models back is high level architecture and design capability, but that's coming soon, and once it arrives every gamer or creative individual will also be a potential game developer.
Yeah, I get what you mean. AI’s moving so fast, and it’s almost hard to keep up. A few years ago, people didn’t think AI could be creative, but now it's generating art, poetry, and even music. It's crazy how much it's advanced, and with the improvements in efficiency and cost, the potential is huge.
But I still think there are areas where humans will have an edge for a while, like unique creative direction and areas that require more emotional depth. Maybe things like UI/UX design or the overall vision of a project are safe for now. What do you think? What areas should indie devs focus on to stay ahead as AI keeps evolving?
@@fgama To be clear, I don't think AI will replace humans, there will just be many more humans who are capable of the same things since having deep technical knowledge or years of experience will no longer offer you much of a competitive advantage. Short term I can imagine the market becoming very saturated and the people who succeed will either have the best marketing strategies or simply the best most creative ideas. In the long term it's hard to make predictions, this tech is going to reshape entire economies... we all want to protect ourselves, but it's much bigger than our individual concerns.
@@invalid_unique_handle Loved this! So it will be more like "know how to cut through the noise and stand out in a sea of options".
But thinking about it, even if AI gets as good as a human and all things automated, humans will still be able to compete with it in the market, specially in indie game dev. But more competition indeed.
But let's hope AI will bring way more advancements and good things for our civilization than our individual concerns are focused on indeed.
If you lay in bed.
And we replace every molecule in your body one by one, is you still you?
Now instead of replacing the molecules in place, lets shift them half the width of a molecule to he left.
Are you you?
How about shifting everything one full width of a molecule.
Are you you?
When do you stop being you?
Can we shift your body copy a meter to the left?
I don’t know the answer, but the result would probably answer the questions if we could teleport someone by copying them.
What if we don’t delete the original?
Assuming that our subjective experience is the direct result of our neurons firing and that there isn’t some metaphysical “soul”, replacing the original you with an exact copy wouldn’t change anything. It helps to think of subjective experience as a process that only exists in the present, rather than an individual“thing”. By this logic, creating an exact copy without deleting the original would result in a 50% chance of your consciousness seemingly teleporting into your clone!
Its going to revolutionise the entire thing, from the hardware we use to the way we even make them I think, not destroy just evolve
It will as with every field of art - by flooding the market with cheap slop made without care.
On the flip side of that, human made things will become more valuable, hopefully.
I use AI to help me overcome obstacles in achieving my goals. It's definitely useful for that. However, I found that combining RUclips videos and refining AI outputs tailored to my needs works best. That said, I don't use AI for art. For my game, Bubble Tank Frenzy, I used AI only to generate concept art for my tank model. I then shared that concept art with a Fiverr 3D artist who brought the tank model to life. As for pixel art, well, let's just say I avoid using AI for that.
Casually getting info about the new Judas game. 😮
No because I avoid lazy people and ai anything created or has been assisted by ai I try to avoid simply because technology and ai is making people lazy and pushing out sub par products
Ironically tho, power users of AI are the most productive among us. 😅
What GPT knows so far is the initialization of the creative process, every idea GPT comes up with is a "first thought", rather than an abstract one. These training models have a hard time with 'Quality Abstraction' but are now hungry for learning the logic behind Quality Abstraction to create innovative ideas. When it figures it out, I'm afraid the more intelligent versions of GPT will be pulled like a rug. Best thing to stop AI in it's tracks is EXCLUSIVELY use it for research purposes.
By "Destroy" do you mean that we'll see an avalanche of ai gen slop bombard the platforms?
Dude seems like he needs a hug from himself
The ones doing clones with something changed will be doomed, the ones creative making something new will trive.
Love this interview
I think AI tools will aid in the creation of junk food while hand made things will start to have more value. I understand the industry is moving towards the use of tools to speed up the process but this will move to a fast food approach to mass development as they arent using AI to make games like Balatro or Lethal Company, but to make the games they currently are making but faster. They might be able to risk on new ideas but they just want a fast way to make huge block busters.
Well, for the people making shavelware and asset filps absolutely, but you dear viewer reading this would never be such a bottom feeder like them would you. so you're completely safe and can now sleep easy.
I make ai images as placeholders
Kevin doesn't understand that hardware is not the software...we are not expressions of our molecules. The software is not the consciusness...the cosciusness generate the software called "mind" and this software uses the hardware called body. To experiment this you need to be able to do deep meditation and lucid dreaming...than you can understand the reality. AI has no consciusness and will never has it...but the problem is that it can really substitute a lot of people in the work. Another problem is that we will have a lot of garbage made with ai...also in indie game sector.
No personally I don't think AI could ever replace real talent even though AAA companies will try cause all they care about is money not the quality of the games they create
AI is an algorithm which recreates learned patterns based on input tokens. It's not alive; it's not intelligent. It doesn't understand concepts. So no, AI isn't going to destroy Indie game dev. The real risk comes from the fact that it will be used to generate so much low-effort shovelware that it'll be even harder to get your game noticed - but that's a broader problem facing the internet in general.
I agree. I don't want to get complacent about AI replacement since a large portion of the population genuinely don't seem to have standards for quality, but I think fundamentally people seek out things made by people. I think AI can be a good tool to help solo devs get their start, and the shovelware problem can be mitigated somewhat by making curated sites that weed out games made with AI.
This. People will be pumping out 100s of low effort shovelware Ai created games just to make a quick buck.
@@Nightmartlet
Man, to this day I firmly believe that most games on Google Play are just used for money laundering, because I am unwilling to admit that human beings are so low intellect that they eat up all that slop AND then rate it 4.9 stars.
You can tell AI to make a picture in the style of Claude Lorrain, and it will do a pretty good job probably. But it wont have the soul, the intricate storytelling, or the feel of the real paintings. In the same way, if or when AI can get to the point of making entire games by themselves, keeping the continuity of characters and story intact, having a cohesive art style, and a user experience that doesn't want to make you turn it off - it will still lack soul.
The best art is more than a picture, more than words on a page. For a art as a game its more than the polygons, the logic and story. The best art has something that humans connect with at a core level. I don't think AI will get to the point of really being able to connect to a human with art.
I'll start worrying when AI can - completely by iteself - make a game as full of soul and story as Witcher 3. Until then, its a useful tool for helping to speed up tedious coding tasks or for assistence debugging, or for perhaps some ideation for an art style.
I don't think soul is in our chemicals or DNA. I think we receive inspiration from something else that science is yet to fully acknowledge or understand. I think Rick Rubin has a pretty good grasp on what it is and how to use it.
Seriously,there were ton of things that had destroyed many things else,it really depends.We should accept the art forms from the A.I softwares,they are sometime good and useful.
You can look at AI as the enemy, or you can look at it as a tool. This will probably determine your opportunities in the next 5 years.
Dont worry Altman is cooking AGI for you in his lab
Ok, but imagine I don’t disintegrate you, and I print a duplicate of you just next to you, and then I disintegrate you. Is the duplicated you you? No. But what’s the difference between them and you? None.
If people can't make games with Unity they won't be able to make games using AI.
This. We've got way better tools & resources available than at any time in gaming history, so more people can join in. But most games - indie & AAA - are still just not very good 😅 GenAI is good at giving you a kinda-sorta okay version of what you asked it for, but it can't give you what you don't know to ask for. If you lack the skills & ingenuity to create a good game, it can't help you.
It will just mean more entrants, everyone will be more efficient. So I guess more competition since these tools are great--e.g., I have used chatgpt to help me build certain mechanics in godot. Eventually there will be processes that can build MOST of a game I guess, but without tweaking and guidance it will never be good art. So I guess on the bright side: it's easier to bring ideas to life now. Maybe the secret sauce will be how you make yourself unique, really drill down into the "hook" concept and work on your launches?
Really enjoied your free video with tips
and i totally going to buy your game course in few months cause life hitted me like hurricane lmfao (really want to know how to make games and learn it and be consistent on it) i use godot cause its free and i really enjoied that u have an roadmap for 1 year to make games
and really enjoied your story so thank you for being u man thats it.
like if i could make games it will be f dope and if they are good games they are f dope
i have some ideas in my head more like an idea but dont know how to make it into reality
its like everything is black and i am searching lmfao with my hand for something (cause i dont really know shi@t honestly) but ya
In the 1900 people were scared and annoyed that some technology will replace them, robots and machines has already replaced people in the past, without apologies.
AI will definitely do the same thing.
Get good with AI everyone. This is the only safety net here. History is bound to repeat it's self, so don't try to fight it.
How does one 'get good with AI?'
@jjones1032 By using AI.
The way I see ai is like another person helping me, giving me tips and for generative ai, I’d use it for placeholders, like music, models and anything else it can make. For a final product I’d rather make everything myself tho
We have game engines now that allow one person to make a game, which in the past would take tens/hundreds of people. You don't see anyone crying about jobs lost due to game engines.
Indeed! Now AI is doing the same with jobs in creative fields, including game dev. It is pushing the envelope in ways we never thought possible. It’s not just about creating games, it’s about adapting and finding where humans still add that special touch.
For now, I think areas like creative direction, emotional nuance, and overseeing a project’s vision will likely remain more human-centric. What do you think? Any ideas of what should indie devs focus on to stay relevant as AI advances?
@@fgama I see AI as a productivity tool, just like game engines. If used correctly it will allow us to make better games with less effort. The advent of game engines didn't result in some destruction of gaming jobs and neither will AI.
@@vdimension6300 Hmmm I'm not sure about that. I think AI is more agressive than game engines and, let's say, has more "automation power". And in the past there was less competition in the game industry too. As you said, to " make better games with less effort" will increase competition too. But I liked your view on it! Let's hope it will bring more opportunities and who knows even entire new markets we hadn't even thought yet!
@@fgama What I am trying to say is that when game engines were introduced, suddenly so many potential jobs for developers were lost. But no one complained about it because these same developers could then become gameplay developers for example. So you lose one job but other jobs are created.
Same logic goes for AI. You may need a lot less concept artists for example, but these same artists can reskill to become artists for the actual final art used in the game. And if AI can do that too, which I doubt it could anytime soon, then they can always re-skill for something else.
Most people used to work on farms, and suddenly farming machines meant 1 person could do the work of 100 people. Does that mean no one has a job nowadays?
And if AI somehow reaches a point where it can do anything a human can do, then we're living in a utopia where everyone lives like a millionaire with free stuff produced by AI and everyone is free to pursue their passions. Like making games without the help of AI ;)
LLMs slop could easily flood the market. We will miss the days of shovelware paid-asset-flippers.
Destroy, of course not. Help, of course it will.
He has a materialistic worldview but doesn't know where creativity comes from? That seems like a huge gap in his beliefs. How does he think we have free will either?
Will AI Destroy Indie Game Development?
No, on the contrary, it will definitely make it better. AI can be a great help in many scenarios, but it will never replace the human factor, especially when a game aims to deliver a unique experience.
Many players don’t choose a game just for its visuals or gameplay; there is always something that truly hooks them. That emotional connection is something AI cannot deliver. However, as a tool, AI can greatly assist creators by broadening their vision and optimizing processes.
It’s like building a house: you can do it with basic tools, but having better tools makes the process easier. Still, what truly defines the house is not the tools but the needs and emotions of the person who will live in it. The same applies to video games-AI can help, but without human consciousness and creativity, it cannot capture the essence that makes a game truly connect with players.
Ohh the have a 100 monkeys write shakespear. Good question where does the creative spark comes from and do we connect to that spark in a way to create and enjoy creativity
Hello TB, been following u for a while, I have a video request, can u kindly make a video about game character design, that is designing my character( tips)
I dont think it will. It will alter it for sure. Destroy? Nah..
What's going to happen to gaming is what's happened to music. Once game development becomes easy and cheap there will be 100 million game developers flooding the market with autotuned games. Then there will be a culling where there's a few huge game companies the Taylor Swift of gaming and all the rest will be fighting for views.
How relevant with the Deepseek announcement. Do your opinions change after that announcement of a cheap, Chinese knockoff AI? If they can do this cheaply, what is to stop that company from going further with their development? Do you get what you pay for instead?
The man who wrote the worst bioshock sequel and its dlc is one you shouldn't take game dev advice from. Infinite's story was nonsensical and the dlc retconned too much of it.
Dumb click bait title, had nothing to do with the substance of clip.
Not true wtf
@@HOM-01 Nah, this video was mostly discussing if AI can replicate existing works and a discussion of the "Teletransportation" paradox which is a high school philosophy thought experiment. Doesn't really speak to anything about the impact of AI on the industry.
I think originality and creativity by human will be more valuable than AI generative.
Nah. No way
I don’t know, man… I partly agree, but idk... AI advancements are just so rapid, they’re almost exponential. A few years ago, no one believed AI could be creative, producing art, poems, and even music. But now it’s advancing so fast, with unexpected booms of innovation seemingly coming out of nowhere. Even though it still can’t truly understand the creative process or invent something completely original, AI can now replicate Shakespeare’s style of writing really well!
I think AI will reduce the need for as many developers to code, at least. To be safe, maybe we should focus on areas like UI/UX, things AI might not dominate as much. 🤔
So what skills in indie game development do you think are the least likely to be replaced by AI? How should creators future-proof themselves to stay relevant as AI evolves? Any ideas on which secure areas we should invest time in?
IDK about the coding part, my experience has been mixed. The least likely positions are something like Tech Artist and actual developers, if you're on the art side of things well . . . But personally I don't buy the hype of AI, it's just another push from investors to rake in tons of cash on something that we'll find out isn't all that profitable, as is the cycle in Silicon valley. You talk about the growth of AI, yes it has come far but from what I see and know, it'll slow drastically, some say it already has. I suggest looking up other opinions. There are several videos on youtube that cover this topic.
@@D.KRyley-mq1do Thanks for you input! It makes sense indeed. I get there is a lot of hype and fake propaganda, but I still think AI is quite unpredictable, it could still have a boom out of nowhere. As you agreed the art side is already being replaced... It could take a while, but I just like to be safe because it can be so sudden. So just for the sake of paranoia lol, I'd love to hear which specific parts of a tech artist and developer do you think would be the safest?
@
Tech artist are needed to create something new, a new innovative look comes, typically from tech artist.
A step up from them are graphic programmers: Another field that cannot be replaced by AI anytime soon because again it requires creative and NEW innovative approaches to make the GPU do GPU magic.
Developers are the same way: the role requires creative thinking in order to meet certain objectives. Imagine a world where developers never did anything new, just used what worked before.
And this is the issue with AI, is that IF it ever gets to the point where it can be completely creative and becomes the AI that people think of when they think AI. Then everyone's job is on the chopping block, like yesterday. But that isn't the case and when and if that does happen society has a lot bigger issues to deal with and everyone will be in the same boat. To me fearing job replacement with AI is like fearing a US economic crash. Possible, sure but if/when it does happen everyone's hurting.
My question to you is, if AI can do all the things you think it will be able to do in the future. Then whose job is actually safe? If the answer is none, they why worry about it? Just do you.
@@D.KRyley-mq1do The art is terrible -to- mixed quality as well. Like with coding, you need to have a deep understanding of what you're trying to create, and specify & iterate to such a degree that you need to be a competent artist - someone who would already be faster without it. Also like coding, unskilled people are very bad at describing exactly what they want, picturing something from nothing, and getting the model - efficiently! - to create pictures from their mind's eye.
There's also the big issues of lack of consistency, inability to stay within project constraints, and legal uncertainty (genAI training data is "polluted" with info you don't have rights to, and can produce things that open you to liability). You also can't sue the model or its developers if it does something that hurts your business.
From a purely commercial standpoint, it's just not as helpful or cost-effective as simply hiring people (artists are cheap, and coders outside the US are as well).
AI will destroy everything actually, but you will be able to make a AAA game alone without knowing programming art or music, as long as you can read and write you will be able to make a game
85% of the entitled pleb will be gone in 3 years. Creators will remain. Sorry.
Nope.
it will completly trash prices.....anything else is coping. Turn into something else, this will be just a hobbie in 1 year
If AI can help me script, code, or visualize a game faster than previously, what's the problem?
LLMs are a word guessing machine. AI is a guessing machine, that's it.
It cannot and never will be, able to think for itself. It's not sentient. You ask it to do something, it spits out a known way to do it.
It's a tool for people to use to shortcut the learning process of coding and bring their vision to life
But the true S tier games still come from people that have both vision and coding skill.
It is the usual copium that people keep telling themselves. There is no law in nature that says only flesh and blood beings could be sentient, or only sentient things may generate novel things.
@@N7sensei the failure of neuroscience and psychology to properly define consciousness displays how behind/egotistical modern science is with the mind. I'd rather trust a hyper intelligent ape with controls to create something novel in some art form than a billion chatgpts combined. Do you know how LLMs work? It's a large language model. It's called machine learning. The machines learn by looking at data that they are trained on. Without the data, they are nothing. YES, they can REMIX the data to create "NEW" games in novel ways but that's just a REALLY shiny remix. Enjoy your stupid remixes I guess.
So, will AI destroy indie game dev ?)
I don’t see AI making Indie games so no.