The Impact of AI on Game Dev Jobs. Open to Work Crisis(darkounity.com) |
The Impact of AI on Game Dev Jobs. Open to Work Crisis(darkounity.com) |
Also, the Open to Work banner has the stink of desperation. Highly recommend disabling it. It's like dating. Act casual.
I have little doubt that if I had an opportunity to use Claude to do my CS homework, I would have used it. It seems that the curriculum should assume that college kids are going to use the latest agents and dramatically increase how hard the homework is.
It may as well say: "need work, bad"
Today... it's Roblox vs everything else. Steam has its data at https://store.steampowered.com/charts/ - in the past 3 days, 45 million peak online at once.
Roblox has 150 million daily active users and 380 million monthly active users. https://brands.roblox.com/metrics-insights. Different numbers. Back in August it hit 47.3 M concurrent users (Steam's record at that time was 41.2 M).
As those gammers are growing up... they're not switching to other platforms.
Blaming this on AI doesn't feel like most of the story.
Gamers are dying out(*) - Belluar News https://youtu.be/_80DVbedWiI
Partly it's because they're easier to get into. It works well on a Mac or even on a phone. Lie in bed, do your three missions, sleep.
Partly because I can play with the kids. No extra account purchases, no teabagging.
I think game devs assume that they're just different audiences, but not really. Just because I like 18+ rated games sometimes, it doesn't mean I wouldn't pay a 6+ rated game most of the time. Who knows, we might see a Civ equivalent on Roblox in time. Roblox is the metaverse that Meta saw a decade away but couldn't capture market share for.
Also how believable are both Steam and Roblox's numbers with regard to bot accounts? I have no idea about this, I'm biased towards Steam but based on the HN comments every time Roblox comes up I get the vibes that they are a comparatively less ethical company that might be willing to push the boundaries of ignoring bots for the sake of better metrics to impress investors. Also this isn't data at all but the steam games I play wouldn't benefit from bots, but I could see players of some of the silly little games my kids play in Roblox benefit from having some bot accounts to help you.
So wouldn't the above numbers indicate that you have a growing younger gamer population (e.g. in Roblox), and an older one in Steam? And that it would be reasonable to think that many of the current Roblox gamers would transition across to games in platforms like steam, but they cater to a different need or desire in the coming decades?
Or does Roblox have everything that games offer on Steam already? Can I get my Sekiro experience there, or Star Sector, or Civ, or Witcher?
There's no denying that Roblox is immensely popular, but maybe it's OK if Steam has more niche stuff in the long run? There's still big money to be made, especially for small developers.
And also aren't we comparing overall platforms to individual game performance? To compare _all_ of roblox against a single game (BF6) doesn't seem right, no? Shouldn't we be comparing the most popular Roblox experience instead (whatever that may be)?
Those games are enormously popular and might need some context of their own let's continue: Battlefield 6 has had at most 774k concurrent players, and peaked at 22.5M MAU
so Roblox is about 30% more popular than Minecraft and candy, or a good 19 times more popular than battlefield 6.
Someone is getting a new yacht.
This is the first time I heard of Roblox. And it's supposedly this huge phenomenon?
I guess this is what getting old feels like.
https://www.takeaway-reality.com/post/roblox-demographics-st...
There are 151.5 million daily active users on Roblox. (70% growth YoY)
44% of Roblox users are over 17 years old. 56% are younger.
APAC has the most users (29.5%), followed closely by US & Canada (28%). Europe is the third largest market with 23%, while the rest of the world has 19.4%
As of February 2025, the platform has reported an average of 85.3 million daily active users. According to the company, their monthly player base includes half of all American children under the age of 16.I wonder what happened to make it so popular in the present day. Was it incremental improvement, or a better fit with modern devices/culture; maybe simply being free to play?
Very similar to how markets worked before central banks emerged. So someone, say a King, would over print cash for whatever random shit his self important 3 inch chimp brain came up with, like War. But at the market place there was nothing to signal random ppl now had more cash in their pocket. How much cash they had in their pocket got decoupled from any work they actually did. So strange things would start happening in the market - prices would jump, factories and farms couldnt produces at the same rate cash was printed, banks would lend too much, people would blame each other or outsiders because no one had info about how much cash was being printed and a mechanism to adjust its supply. Bankruns, asset bubbles, market crashes were the norm. Who looked rich had more to do with how close they were to the money printer.
> He has a huge influence on the developer community, and he is deceiving his viewers about what is going on.
The sub bubbles of development are so crazy to hear about. Idk if I ever interact with anyone that gets their development world view from people on YouTube.
Look up FRED data in Interest Rates, SWE job postings over time and then look at the stock price of one of these companies claiming “AI layoffs” over the same time frame (eg. Block). Then read reports of 0 discernible AI impact to US GDP in 2025…
Really, interest rates? Not the change in the law that made it illegal to treat software developer salaries as a business expense?
I know that learning averse programmers exist, but even pre AI, many people in this position would just learn some python
I wonder if author left this as some kind of Easter egg or it's just another life is a simulation moment. In this paragraph he used semicolon and parentheses multiple times. After that only once.
Anyway love the article. It's nice to hear from time to time that one is not becoming obsolete that quick.
This is the reason why you can hardly trust AI companies to not train on your code. If you care about privacy, I think it's best to use Antrophic models via a third-party provider like GitHub Copilot (business) or Amazon Bedrock.
I play pokemonshowdown.com sometimes. You can message me there or mail me your username and we can decide a time to play :-)
My username on pokemonshowdown.com is smilenowpleas
You’re pretty much spot on.
Isn't this risk also present with offshorting blue collar jobs? I agree that both are bad ideas, but the capitalist class didn't care before and isn't exactly well known for it's long-term thinking.
We need less folks ringing alarm bells without guidance and more folks offering help in troubling times. This, is not helpful.
There are problems, some are very visible, some are present but not very visible, and some are imaginary.
There is a growing discontent like I have not seen in my lifetime, and I don't think much of it can be characterised as unjustified. The problem is the narratives that people are forming about those problems are mostly all wrong. This is easily provable by showing how mutually exclusive all of the narratives are, the correctness of any one of them would invalidate many more.
I don't don't have the expertise or data to conclusively show what the cause of any particular aspect. To make a guess, I would look to history and note that those who blame the new thing are usually wrong, and those who warn about actions being done without regard to the consequences are usually right.
People are angry and want to understand what is causing the problems, and it seems the unscrupulous have found that an answer to their questions is effective, even if it is a fiction.
> I don't don't have the expertise or data to conclusively show what the cause of any particular aspect.
For whatever it's worth, I think we're on the same page here. My personal observations shore up those same assertions, and they're very much systemic in nature rather than singular topic like prior grievances.
AI may just be the most recent inflection point bubbling more of this discontent to the surface, but said discontent has been gradually rising to a boil for decades.
1) Growth is gone and all expansion narratives have failed (VR/XR, NFT, etc) with no rescue in sight. That’s a death sentence in capitalism.
2) Demographic headwind - not only from age but also habits - short form video is attention intensive, highly addictive and doesn’t co-exist like a spotify or youtube clip in the background. And it’s wildly successful. New generations are not automatically gamers anymore, many just watch.
3) there’s plenty of cost competitive attention targets now all competing. A video subscription is easily competitive with most gaming options on price. It used to be games are THE cost effective way for people to be entertained.
4) The pacemaker is dead. Nvidia used to drive the industry forward with GPU expansion - new graphics capabilities offering better games and fantasies. By RDR2 this stalled, you can’t sell on “breathtaking graphics” driven by technicals anymore, all new experiences have to be gameplay based which is very very hard. 4k was already a bridge to far.
Worse, Nvidia abandoning the industry again (as with crypto) and the AI boom is massively increasing cost of entry for games as entertainment compared to other activities, cost of business (game studios do need a lot of GPU), kills attempts to shift the business model to streaming where you’d have expanded the audience by reducing hardware entry point (not a good use of GPU), murders new consoles whose marketing campaigns always provided strong market rejuvenation.
5) Competition with its own supply - especially because hardware stalled, 10-15 year old games look perfectly alright and run great. They cost 3.99$ in a sale and compete on time with new titles. Successful GaaS titles like PubG, Minecraft, Fortnite, Dota, Lol go strong and cannibalize a vast amount of attention.
Because Valve is privately owned and protective of their market, nobody can buy up the supply and remove it, which kills the proposition of game pass models.
6) There’s too much supply of games to allow all the companies in the space to make a living. It’s gonna get a lot worse with AI, which threatens to kill the artificially high entry cost to AAA by scaling down both talent and technical moats is another reason for investors to run away.
7) When supply goes up, GTM costs become the primary business problem. We already saw this in mobile which has zero barrier of entry, chinese and vietnamese shops will budget build your game to specs.
But like sneakers, anyone able to make a game doesn’t sell them because supply side auction costs go up trying to reach eyeballs.
Facebook killed and ate mobile gaming because app install ads became the primary way of user acquisition - the infamous fake game video trend was the result: The video sells the game, not the game itself and an entire industry jumped up making attention optimized videos to sell to game developers. CAC is extracting all profit from the industry and VCs abandoned it because the extraction makes their financial goals impossible.
Failure to enforce truth in advertising regulation means competition happens more and more on fakes and grifts which favors the grifter.
With AI commoditisation is going to move up the stack into the rest of gaming.
And crucially, any narrative of more games or cheaper games runs into simple physics - we maxed the available attention which is colonized by companies who have genetically engineered themselves to maximum addiction. Without new time (4 day work week), it’s a brutal red ocean game companies cannot hope to win in, given that the platforms own the attention algorithms.
More supply in a red ocean means less money for creators and publishers and their shareholders and more for the platform”, even if you cut creatives with AI
8) There is no credible future vision or narrative for gaming.
Yes AI will allow new experiences but nobody knows how they will look, IP protection is dead so even if you create one, there’s no way to do it at current economics.
Other uses like character.ai, again, A:B tested to maximum addictive/ sycophantic behavior seem like attention headwind.
All these PLUS AI investment having “better” bets (like betting on OpenAI to become a state sanctioned monopolist getting paid back by taxpayer money) means that there is no money, no lifeblood for the industry.
It’s going to burn down to a tiny shell before something can sprout from the ashes, in the tiny plot the attention platforms allow it to take.
AI adoption across the industry is anemic in effect. It’s impossible to build native AI studios because the ground and definition of native is shifting too fast. R&D risks and constantly shifting capabilities in the frontier are extreme, you can’t build a realistic financial projection. And even the cost savings that are there, theoretically, like voice over or asset generation are poison to an overwhelmingly negative to AI audience.
Overhiring is often cited as a reason and simultaneous cope to indicate temporary correction but games are multi year products and only a small segment of employers (platforms) engaged in it - there’s little evidence to support it.
ZIRP is also often paraded around, but ZIRP is short term and certainly a reason - but the tech industry, which had the same challenges, shows why that’s a cope and why the lack of narrative is terminal: There is no limit to the amount of investment you can collect on a good narrative. The industry has none.
Yes, AI narrative is killing the industry by choking and diverting oxygen, but technically AI has, for all intends and purposes, zero effect on gaming yet.
Source: Industry veteran in AAA and Big Tech.
But it is legitimately incredible what some people created in it. I remember playing a very good/accurate counter strike source clone in roblox. Felt exactly like playing a source engine game. I doubt that it was made by a kid.
This is a very "if you exclude the best and most robust parts of the economy, everything looks terrible!" sort of comment haha
services industries don't contract nearly as much as manufacturing during interest rate rises, see 1980-82 recessionary period
everyone in finance correctly regards entertainment industry as more robust during high interest rates (and the macro environment high rates respond to), that doesn't mean buying Disney is a good trade (it's not). two separate issues.
They won't be fit to work as body builders, sure, but presumably that's not what they were going for when they strapped on the power armour.
Same as CS graduates aren't going to enter a work force that writes code by hand, and shouldn't expect to. The job market requires power armour operators, not muscle heads.
Professional programming without AI assistance is a thing of the past. Much like stablehands or squires or farriers.
You can still do it as a hobby though. You know, for fun. If you want to. It's like knitting!
The question is, what are you trying to learn? There are still going to be problems "power armor" can't solve, like what to move where.
Ideally, the people operating the large powerful vehicle are in fact trained in how to use it safely, because trucks (and power armor, and LLMs) can do a lot of damage if used incorrectly
- everything Roblox says is true
- a company that spent a decade throwing shit at the wall until something stuck, that seemingly has absolutely, utterly zero cultural impact, with a huge, inauditable non-US audience, where every game that any adult has tried is absolutely trash, that has tons of employees but seemingly not a single one has gone onto found a great game studio or do anything interesting in media unlike every other successful game studios, that Apple could ban for violating the mini-apps rule in an afternoon and hence ruin their entire business with a stroke of a pen... is overstating things.