here is your hill, enjoy dying on it

studios refusing to use AI will be beaten by those who do -faster content -more content -lower cost of dev imagine a farmer refusing to use the plow because it took jobs away from hired hands

Per my tweet on Oct 9th:

Hey remember a month ago when we were told AI was gonna kill us all? Seems we have moved on, kind of sus for something that’s supposed to kill us all. Hmm.

guess what, no one bought it

Here are the current top earning Roblox games. Nearly all use both AI coding + AI asset generation. In some cases these games are entirely AI assets, not just one. The top 20 games on Roblox collectively earn $8M A DAY. There are hundreds of millions of people playing these games, and millions spending daily. So the answer to who is paying for AI assets is very clear: the generation replacing us as gamers.

Per Shawn Layden's comments a few years ago.

Growing a single California almond requires up to 10,000 conversational AI prompts to match its water footprint.

If you produced cartoons for Nickelodeon or Cartoon Network who paid you for your work, and if an AI scanned that image without license, it was Nickelodeon being stolen from rather than the artist. As with many products, Nickelodeon is able to license its IP to others without the consent of the artists who originally worked on them, though Nick may pay royalties to them depending on how the contracts were done. Nickelodeon is now owned by Skydance who's begun an IP program for licensing its work to AI companies. AI using Skydance-licensed art is no longer theft, it is a revenue stream! The theft argument will slowly die as these deals become commonplace.

dont worry keep on using your abacus one day you'll solve that math problem

AI is a tsunami coming for you. Ride it or die.

Epic, by far one of the best technical leaders in games, is saying no hand written code. If you're not doing the same you will be left behind. It is over for non-AI workflows.

Kudos, this piece is incredible. Here's a studio who treats its workers well; the piece cannot name an actual grievance with DoubleFine itself. There is not a complaint about pay, there is not a complaint about actual crunch (there is just a note to a 2019 concern that there might be crunch), there are no poor working conditions. As noted in the follow up interview with Tim, the union rules were written when people were making railroads. You know, work where their lives were on the line. No lives on the line here, just workers with a grand entitlement complex. The desire for the union at DoubleFine was pointless, a snake eating its own tail at a studio who was cut for losing money, and who lives on borrowed time still today in Tim's own words! I wish the studio luck, but not because of its unions. All of games faces the AI tsunami. There are almost no people wanting to write checks to US games studios unless they're a small team planning to do all the work with AI. No one NEEDS to write code anymore, and those who do are eating the burn rate at a studio that could actually put out products with that programmer cash instead. Art is swiftly going that way as well, while younger audiences are satisfied with Roblox and cheaper international dev will swallow the rest of the dollars going toward studio-led games. The old way is over. The cost structure does not work. The barbarians have already knocked down the walls of Helm's Deep but sure, attack DoubleFine over unions over a studio that was a dead man walking and appears to have only saved by the skin of its teeth by Tim's leadership. Great use of time and energy.

If you are still writing code for a game by hand you are driving a horse and buggy and every other studio is leaving you in the dust with a supercar. Remember, time is money.

You have made an image that looks like a standard conspiracy theory board. If that was not your intent, then I apologize.

Pinkerton at 30! Has there ever been a sophomore album in history that flipped from so hated to so beloved by a fan base? From last week’s Weezer concert in Brooklyn.

You can optimize your game for profit or for social impact, but you cannot do both. In the mid 2000s there was a debate in the industry on video game box covers. The box covers all looked the same: generic white male protagonist / same stance with a gun. There was a reason the marketing departments did these box covers: they had an audience they were chasing, and the market research said that this style was the best selling. Their job was to sell as many units as possible. We started to see criticism in the press and forums about the box covers being the same. At some point in the early 2010s/mid 2010s there was a discussion in the marketing departments about the need to move away from this. However, the need to move away from this wasn't based on market research for what the existing audience wanted, it was based on a desire to expand the audience and improve social impact. By this time many more girls were playing games, many more regions around the world were playing games, but not necessarily these games. Tomb Raider's audience didn't suddenly expand because we changed Lara Croft's look from large pixellated breasts to a more sleek physique. The market did in fact expand, greatly. Many of the new gamers (gender, region) were on phones, and were playing different kinds of games than AAA packaged disc console products. Facebook's first canvas games famously brought in a lot of soccer moms to play Farmville. So for AAA, packaged disc console games, we entered into a bit of a conundrum where we thought market expansion was happening to us and we could forcefully expand this market through marketing initiatives that moved away from cliches. I do not believe that this was successful. We have seen more girl gamers come into console in the last 15 years than the prior 15 (console is now only 59% male), but CoD is still famously a male base (67% male), and it's hard to say that the core audience has overwhelmingly shifted for packaged AAA, and regional concentration remains massively dominated by US followed by Europe and Japan. The rest of the world (China, SE Asia, LATAM, Africa) all went mobile. Marketing departments for consoles working on shooters would be better served doing a deep analysis of who their core audience is and what they want, rather than trying to expand again. That fight is over. If there was an expanded audience for console shooters specifically, Fortnite already won it.

If you are bashing Ben Affleck for using AI when he filmed his own footage for training, you are masks off when it comes to the goal here: "theft" and "ethics" of reuse were never legitimate criticisms. The problem is clear: are upset that AI will take jobs from them, because it does VFX faster and cheaper than humans. This has always been the way with technology. Switchboards replaced operators. Automated elevators replaced Elevator operators. Factory robots replaced many people doing things by hand, etc. You cannot stand in front of a tsunami and try to punch it to stop it from bowling over you. You cannot successfully fight the battle against AI taking Hollywood jobs, it is already lost. You will need to learn how to use AI and train on something new or risk drowning in your own ignorance and self-pity. Workers in these industries are not coal miners where the mine is the only work in town. They just don't want to learn something new or change trades despite having lots of warning that the change is coming, faster than people think. Train now with AI. Soon. Before the pink slip. It will give you an advantage in the next role.

Same issue with games. Anyone operating with a huge staff now is operating with an antiquated pipeline. Programming, animation, art, loc. AI does it for almost nothing. Per below. Huge disruption the next few years. People in the C-Suite don't know what to do with the team sizes they have now. Working on a game with AI, I can have it redo something 100 times, it will never complain and it will do it from 4am to 3:59am the next day.

Try now.

Starting an AI Games Festival. Consists of every game made in 2027 because everyone uses Claude or Codex now. If during this festival we discover a game uses human generated content we will have no issue with that. Literally all of steam okay.

this is the bright future, it just takes a while to drive there

The problem with this kind of reporting is the following, and I see it a whole bunch: Journalist has a source in mid-level management. Source give's source's opinion on situation. Source may or may not have insight into actual management's opinion. But source's opinion is taken with a level of legitimacy-- meaning that it represents the company's POV-- that isn't actually relevant. When Tom claims that sources tell him that the preorders ring alarm bells, do we care about the alarm bells of manager guy, or do we care about the alarm bells of Sharma, Booty or Ball? The latter, obviously. Their opinion is the opinion of the company's POV, the others is not. E.g. some mid-level manager may think it would be dumb to license a Kojima game and would rather they make more Kilns, but guess what? Their opinion doesn't matter. Booty's opinion matters. The result: Kojima yes, DF no. So when one of those three steps in and says that what Tom's source told him is not true, that is the actual response that matters because it represents management's actual POV, and the source is proven to be irrelevant. This is what happened to Tom, and he got shown up for having some nonsense, unimportant source.

No, Tom's reply is completely irrelevant. Attached is Tom's piece. He says sources tell him that GTA6 preorders are heavily skewed toward PS5, and then cites that there are alarm bells. There are two points here: 1. The skew of the sales 2. The notion that this is a concern So long as the skew of the GTA6 sales are matching console sales, which Ball is saying is true, then there will be no alarm bells. It does not matter whether that skew is 10/90, 50/50, or 90/10. What matters is whether GTA sales are or are not matching the console's market share. If the GTA sales were not matching the console's market share, then concur on alarm bells. But the actual number is irrelevant to the argument of "ringing more alarm bells"; only the ratio match matters.

You can see this on their sheet, where they note their assets are cash and intercompany receivables. You can learn more about transfer pricing here plantemoran.com/explore-our-…

Below, in the screenshot, is what Tom wrote. GTAVI will of course be heavily skewed toward PS5 if PS5 has the marjority marketshare, which based on public data, it has. Alarm bells would be raised if GTA6 sales numbers were significantly lower for Xbox than its market share. Ball is saying that is not true. Xbox has therefore said it is not worried because its sales match its marketshare. Tom is wrong that alarm bells are ringing.

Working all night on animating some models with Opus 5.5 and its token usage is way more efficient than Astra for the same tasks at medium settings.

I can think of a dozen reasons why a business whose new CEO was installed for a turn around would look at business planning, forecasting and demand and say wow you guys did a bad job, see ya.

To my knowledge neither Capcom nor Konami generated massive losses, while many of these studios have. Xbox is not closing or laying off projects with "middling performance." I believe Ninja Theory has been in the red for a very long time.

Let's be clear: You do know what they are doing, you just do not like it. Since what you want is for them to continue to employ people despite lack of performance, and you realize that you cannot say that part out loud because reasonable people will not agree with you, you attack their character instead: you denigrate their decision making ability and imply that they are bewildered and confused. You must imply that you know better than them for this scheme to work, and so we get the below. Xbox leadership are competent and making the right decisions for their business. You do not have better or unique insight into their business and it is absurd to pretend otherwise.

Microsoft: "Our plan is to improve our margins by closing underperforming businesses" Journalists and gamers who do not understand business: "Why they do this? What is their plan?!"

I go by his own words (highlighted below)

1/ The coming AI apocalypse for some parts of the tech stack has precedent. In the 1990s and early 2000s we saw a huge blossoming of game middleware companies, specialist engineers, etc. as every developer tended to have their own engine and it was sometimes easier to license occlusion, lighting, physics and other parts of the stack than build it yourself. Six years ago I shared the below gif showing via GDC booths from 2010 to 2020 how the prior decade saw the consolidation of that entire stack from independent businesses to being subsumed by Unity and Unreal. Hundreds of businesses and engineering jobs were wiped by this transition, but many more games were made as a result of not needing to pay for myriad licenses and hire additional developers just to rebuild parts of the stack you could get elsewhere. And I am seeing the results in posts like the one shared by Dan where people whose manual labor in coding (water, plugins, even niche kinds of AI tools that had done parts of this and are now obsolete) is being replaced by AI that can do the job just as easily. The middleware decimation took a decade; it's likely the AI decimation occurs within two years.

I’m not trashing him. I’m pointing out he’s said two different things. It was a small enough team that he could reassign (per below) yet it was a big enough team that it threatened the studios cash flow which is apparently positive yet not enough to cover a pre production team. You can both point out inconsistencies and also not insult a person. Or at least some people can, you don’t seem able to.

How is it that Physint is both not yet started and hasn’t chosen an engine and yet also he would have to lay a significant number of people off who are working on it? A small prepro team isn’t enough of a major cost to justify the reaction here. A studio doesn’t need to close or run out of cash flow because of a small prepro team. Either there was a big team working on it to justify this reaction or there wasn’t?