This is why game developers are skeptical of Blockchain, the Metaverse, and AI
This is a talk I gave at the Gamescom Congress 2025, in Cologne, Germany. The talk was originally made for an audience of non-gamedevs, such as politicians, researchers, and journalists. It explains to them why game developers are skeptical of technologies like blockchain, metaverse, and AI, by going over their history in the context of the games industry. Below the video is a written version of the talk.
A history lesson
The videogame industry is at the forefront of technology. We build brand new workflows to be super efficient. Let’s take a look at a cool example! With a palm tree and a well:
We can see the palm tree automatically adjusting so it pokes through natural holes in the roof, instead of conflicting with the existing geometry of the well. Organic models naturally intersecting with hard surfaces like this, is very difficult. By automatically generating the tree, we save hundreds or even thousands of hours of manual work for trees all over our games.
But here is a quick question, when do you think this technology was released? Just think of some years, what could it be? Around what time?
Here is the UI.
That already looks a lot older, right? This technology is actually called SpeedTree. The footage is taken from a YouTube video that was released April 2009. That is 16 years ago, since the publishing of this talk in 2025.
So, this talk is a History Lesson. And it’s called ‘Old news, new package: AI, Procedural Generation, UGC, In-Game trading, Blockchain, and the Metaverse.’
First, a little information on who I am. Who is Robin-Yann Storm? This is important, as it ties into the takeaway of the talk.
I have over 15 years of game development experience. I was a Tool Designer at Io-Interactive, working on the Glacier editor, used to build the Hitman games. I was a Tool Designer at Guerrilla Games, working on the Decima editor and engine, used to build the Horizon games. And a Tool Designer at Apple, in California, working on Reality Composer & Reality Converter. I am also a co-inventor on 5 patents, which you can google if you’d like. Nowadays I am a Freelance Product & UX Designer, working as a consultant for various companies.
So, now that we know who I am, the takeaway of this talk is: When you hear something fantastical, please ask a game industry veteran about it. Because If you work on the future of immersion, you have a responsibility to know the history of these technologies, so that you can guarantee the well-being of your users. And there are lessons to be learned from the past, so let’s go through time!
We will start off with the Blockchain around 2016, then the metaverse around 2020, and then AI around 2022.
The Blockchain
When blockchain came to the games industry, a topic that came up often was:
“What if players could trade items between games?”
Yes, that would be incredible. Which is why Steam did it in 2011.
Here is the steam trading UI from back then. We can see two players exchanging items. They have both placed items into the interface and can decide to trade them.
The Steam trading UI, from the TF2 Wiki: https://wiki.teamfortress.com/wiki/Trading#/media/File:Steam_Item_Trading.png
This functionality worked for a lot of games over the years, even back then. For example in Team Fortress 2, Counter-Strike Global Offensive (CSGO), and DOTA 2. Let's say you have a hat in Team Fortress 2. You can trade that for a gun you can use in CSGO. Or trade a gun from CSGO for a courier you can use in DOTA 2. Or if you have a courier in DOTA 2, you can trade that for a hat that you can wear in Team Fortress 2.
This item trading works across Steam games. For example here is my actual Steam inventory. We can go through this list and see how many games actually have items that are tradable. Some are for in game trading cards, but there's loads of games that allow trading of items that are used in-game.
When we look at my Counter-Strike 2 inventory we can also find some ‘Cases’, like for example this Danger Zone case, which has a unique serial number. We can also see it is worth 2 euros, and about 7000 have been sold in the last 24 hours. There is also another case, for operation Hydra. It’s worth much more, 26 euros, and only 155 have been sold in the last 24 hours. If we click on the ‘Sell’ button, we also see a graph of its cost. We see it was worth more earlier this year, and is now worth less. I can also put this ‘Case’ up for sale and decide for how much I want to sell it.
Further in the inventory there is a special item: A souvenir TEC-9. I got this when unboxing a specific ‘Case’ during an eSports event that was held in Cologne in 2016. Because of this it has unique stickers on it for that event. We can also see it in-game. If I open up my inventory, we can see those items, and that TEC-9. We can see the unique stickers on there are for the players of the match during which I had unboxed it. They are nice and shiny. We can also see it equipped on an in-game character.
So I could send this item to others, and sell it to others to make money from it. I can then use that money to buy other items, or even other games entirely. This technology and workflow has existed for a long time. Blockchain would not ‘solve’ this.
Another topic that came up a lot during the Blockchain days was:
“What if players can sell their own items?”
Again: Yes, that would be cool, which is why Steam already did this many years ago. They allowed the sale of fully 3D modelled and textured items, made by players. Here is a big list from the Team Fortress 2 wiki. We can see how many items this really is. It is huge. Players can make hats, weapons and nowadays even animations, which they can then send to Valve, and if Valve approves them, they can appear in loot boxes. The creators then get a share of the money they make.
Creators can make over hundreds of thousands of dollars on items like this. This is A HUGE economy. Valve even had an economist on staff in 2012 to 2013. His name is Yanis Varoufakis.
Image: Olaf Kosinsky (kosinsky.eu) Licence: CC BY-SA 3.0-de
You may know him, as he later became the Minister for Finance in Greece, during the later stages of the Eurodebt crisis. You can still find Yanis’ old Valve economy blogs on his website.
The items on the community marketplace can also be incredibly valuable, for example:
Karambit, Blue Gem, from https://skinflow.gg/blog/karambit-blue-gem
This is a Karambit, Case Hardened. It has a unique texture on it that is set in such a way that the knife edge is entirely blue. This is randomly unique texture setting that can happen when you unbox it, again in CSGO. It is also colloquially known as the Blue Gem. It was once worth over 1.5 million dollars.
This is what it looks like in-game.
Blue Gem, in-game, @zipelCS - Twitter
It’s very blue. There are more items like this. For example an AK-47 that is very blue, and again there are unique stickers on it from a particular year, which makes it more valuable for certain collectors.
Case hardened AK-47, in-game, @zipelCS - Twitter
There are even black markets available for the buying and selling of these items. So you can look them up on separate websites, do payments outside of Steam, and then do the trade itself inside Steam. Especially when it comes to huge transactions, this also avoids the percentage cut that Steam makes on every sale of every item, which is 5% for Steam, and for example 10% for the developer of the game the item is in. (I am not saying you should do that, I am just saying it exists to show you the scope of what has happened in the past.)
It is a huge economy, also with a lot of gambling involved. None of this is blockchain. It has already existed for ages before that. And so the question here is:
Did you know this?
Do not just ask yourself, also ask others when you discuss these topics of item trading. Ask them: “Did you know this?”. Because if they don’t, it means they are missing fundamental knowledge of the games industry. They may not be well versed enough to discuss these systems and workflows.
Another topic that often came up with blockchain was:
“But the players own the NFT items outside of the game!”
While that is true, games work in particular ways that nullify this concept.
There is always a game, and then some kind of game database for accounts, logins, and tracking items to those accounts. So you have the hats on that database, and you can sell them on the marketplace for the game, and the game gets a cut of those sales.
A simple example of a game and a game database, with money flowing in the game database
But if your account gets banned, the hats are lost. This has happened on Steam, where folks get their account banned, and lose hundreds of thousands of dollars worth of unique items, because banned accounts are not allowed to trade their items anymore.
Account banned, or game lost, means the hats become worthless, and you lose all they’re worth.
So in comes the NFT Blockchain. Now we still need a game, and a game database, because that is how games work. Now there are NFT items, essentially receipts, that point to the hats in the database of the game, so that the game can show the right hats on the right players in-game. You can sell those NFTs outside of the game, anytime on the blockchain, for a bunch of money, and nobody gets a cut except for the trade system that you choose to use.
A simple example of an NFT Blockchain setup for a game
But, you can still lose your account. As in, you can get banned, so your hats can still be lost. The game could also cease running, and become abandoned. Which means the NFTs pointing at the hats lose their worth too. Maybe you can still sell them to someone else, but it is up the game’s developer to decide if that is allowed or not. They can readily decide to severely punish you in whatever way they see fit, and forever ban the item in-game.
Game database gone, means NFTs still lose all their worth
So, even with an NFT blockchain, nothing has changed. When someone says “But the players own the NFTs outside of the game!” they show a lack of game development understanding, because the game still has to respect the NFT. And the game can decide not to. The blockchain does not change the end user experience.
Another interesting example is Ubisoft Quartz, where Ubisoft released playable NFTs. Take a look at this trailer:
In the trailer it said it was a new experience, designed for Ubisoft players, which allows you to collect the ‘first, playable and energy-efficient Ubisoft NFTs’. Their name is digits. They are first available with Ghost Recon, a shooter game, like CSGO.
They are playable in-game, like with CSGO
They are unique digital collectibles, like we saw with CSGO
You can use them to complete missions in style, like we saw with CSGO
They are released in limited editions, like in CSGO.
And they say these ‘digits' are unique. Are they? You can see the problem, right? That trailer was released in 2021. A decade after Valve already did this with Steam, without a blockchain. So the question here is, in MBA Lingo: What’s the value-added? Forbes wrote an article about Quartz later saying: Ubisoft Quartz’s Ghost Recon NFTs appear to have made just $400 total.
I think that a regular database, between all their various Ubisoft games, with in-game useable items, would be fine instead, and probably make a bunch of money. Especially with that same 15% cut Steam uses.
So when people ask:
“Why are people in the games industry so against progress?”
The answer is that they’re not against progress at all. They have just seen it before, and what you are doing isn’t progress. So when video game developers tell you your tech is old, not to quality, or mistaken: Ask them why. Ask about the history of what they have seen. They are being critical of the technology, because the games industry is so much at the forefront of technology. Often the games industry was already doing whatever you are suggesting, for a long while.
On a related note, in recent history, there were Steam payment processor issues, where some games cannot be bought or sold anymore due to payment processor limitations. This then also extended to other game markets, like itch.io. We can learn from this that having a secondary market for payments and transfers is helpful to a huge customer base. So please do try to build that, but make sure you learn from the past if you do. Do not repeat the same mistakes, because it can cause that venture to fail.
Another topic that often came up during the blockchain times was
“What if players could use the exact same item between different games?”
Again, that could be neat. There is also an example of the games industry already doing this. Particularly from… can you guess?
That’s right: Steam. In this case the games Team Fortress 2 and Left 4 Dead. The character named Bill has a hat, Bill’s Hat, and eventually they made this able to be worn by characters in Team Fortress 2. That is again, without blockchain. And you can tell the hats look slightly different between one another, because the games have different art styles.
The ‘Bill’s Hat’ on both Bill from Left 4 Dead, and the Soldier from Team Fortress 2
Making the exact same objects easily useable across engines and editors is a big effort in the games industry, the movie VFX industry, and even the automotive and industrial sectors. That effort is called OpenUSD, a 3D scene description API, which was open sourced by Pixar in 2016.
There is even an alliance for OpenUSD, and their mission is to provide standardisation for 3D objects. They have a huge membership, of trillion dollar companies, all trying to make this a reality. But it isn’t reality yet. I know this, because I host the OpenUSD roundtables at GDC, and so I am a part of some of these conversations. If you want to learn more about OpenUSD, I made a different video about it:
So again, the same hat being worn in different games is unrelated to the Blockchain. It is a completely different problem: Asset pipelines.
So when you hear something that sounds fantastical, please ask a game industry veteran about it. Please ask questions like:
“Have you seen this before?”
“Who did it, and when?”
“What problems did they have?”
“What solutions did they find?”
That way you can build on previous knowledge, avoid easy mistakes, and build something that stands the test of time.
Next, let’s take a look at the Metaverse.
Metaverse
Metaverse is a tough term, because the definition of ‘Metaverse’ is up for debate. Let’s go through a couple of definitions, to see what we can learn.
What is a metaverse?
Is it a lot of IP? A lot of intellectual property?
Ready Player One, both the book and the movie, certainly popularized this. With lots of IP available for everyone to play with.
A screenshot from the Ready Player One trailer, showing video game characters from different franchises: Overwatch, and Street Fighter
So, is that the metaverse? If so, Fortnite did it! Here is a video from PattyNest on YouTube.
We can see Spiderman with a gun, and then suddenly Goku walks in with a gun, and then someone gets killed by a kamehameha. This is incredible! A few years ago this would be considered a total fantasy. This much IP, used in these ways, for everyone to play around with? It’s amazing!
But…
It is entirely within their own game. Fortnite, and UEFN. The way I have heard this phrased well is: ‘There is no metaverse yet, but there are IP bubbles’, said by Brendan ‘PlayerUnknown’ Greene at the INDIGO 2025 conference. I think that is a very apt way of phrasing it.
So, what is a metaverse then?
Is it people meeting in digital 3D worlds?
If so, online games did it! Like in World of Warcraft, where sometimes if a person dies in real life, folks hold a funeral for them in-game to pay respect. Or folks meet each other in World of Warcraft, and then get married in-game, and even in real life! And this was released in 2004! I even met my guild leader in real life, whom I used to raid with in the burning crusade, back in 2008.
A wedding crashers video in World of Warcraft
There is also Second Life, released in 2003, that allowed people to meet up in 3D spaces, and did so many more revolutionary things in its time.
By HyacintheLuynes - Own work, CC BY-SA 3.0, https://commons.wikimedia.org/w/index.php?curid=33551991
If you want to go even further back in time, there is also EverQuest, released in 1999.
A very old screenshot from EverQuest, released in 1999
You can tell from the graphics this is quite an older game. We could go even further back in time, but I think you get the point.
So what is a metaverse?
Is it real-life interaction?
If so, PokemonGO did it! I hope you played this when it released in 2016. You would literally bump into random people on the street, and start playing together.
The company that made PokemonGO, Niantic, used to be called Niantic Labs. And they released another game before PokemonGO. It was called Ingress. You would pick a team in Ingress, play on your phone, and go to locations physically in real life. Then you would battle over those locations, and it would create a map spanned out over the real world, shown digitally on your phone. You would meet real people playing this game.
Does that sound familiar? The ingress database was used as a bedrock by Niantic for the PokemonGO database, so that they would have pokestops right away on launch! They took ingress, gave it a coat of paint of an amazingly well liked IP, and it went bananas! Is that a metaverse?
What is a metaverse?
Is it buying digital property?
If so, Final Fantasy XIV (FF14) did it! It was released in 2010, by Square Enix. It is an MMO in the Final Fantasy IP. In-game, there were a finite amount of locations available for housing. When a house went on sale, you would have to literally wait in front of a for sale sign, and bid the highest to be able to get it, overbidding many others.
Final Fantasy XIV - Square Enix
So, finite digital property? That has been done.
So what is a metaverse?
Is it having legs in VR?
If so, VRChat did it.
Just as an aside: The amount of high technology that furries have made, is crazy to think about, and would be another talk all on its own.
So, is there a Metaverse yet? A counterargument that I often hear for this is: “Companies are investing billions into these technologies!”. And again, that is true. I cannot refute that. For example, Facebook, in 2021, renamed itself to Meta. That’s a trillion dollar company! But the question is: When have you been in a metaverse? Have you? It’s been a couple of years now.
It’s important to understand that investment does not equal success. Just because a lot of money is involved, does not mean it has wide appeal. Remember the NFT that was sold for $60 million at an auction. Or Paris Hilton talking about a monkey NFT on The Tonight Show. When have you used or heard about NFTs recently? Investment does not equal success.
And again, I want to be clear, similarly as to with secondary payment markets being helpful if they exist: The combination of these Metaverse elements, plus the required technology, could still happen! We could still get a Metaverse out there, if you learn from the past and do not repeat the same mistakes. So please still build this!
There are also others who know even more about this! For example, Raph Koster gave a fantastic talk called ‘Real Talk About A Real Metaverse’. He goes over the specific technology needed for a Metaverse, and the dates some of them were achieved over the decades. I highly recommend viewing it. Thank you to Celia Hodent for telling me about this talk!
So, when you hear something fantastical, please ask a game industry veteran about it. That goes for the Metaverse.
And now last, but certainly not least…
AI. Artificial Intelligence.
AI, procedural generation, genAI, again the nomenclature is confusing. What is ‘AI’? Is it ChatGPT, and Other LLMs? Or is it Machine Learning, Procedural Content, Simulation, NPCs, Behaviour trees, even just Algorithms? It is important to be clear about what you are talking about, when talking about AI. Because if you are only talking about ChatGPT and LLMs, you are missing the point in game development. All of the others ones are AI too. These have existed for ages, and are awesome.
This naming issue of what ‘AI’ means has been going on for a long while in game development. At the Game Developers Conference, GDC, in 2009 there was a talk from Steve Rabin about defining Game_AI. So this was difficult even back then!
So when someone says “You should use AI, otherwise you will be left behind!” to a game developer, it’s bit like saying: “You should use Algorithms, otherwise you will be left behind!”. Like, yeah, we use technology to speed up work. But that does not always mean having to use LLMs. Especially because more generation does not mean faster game development. We have learned this lesson in the games industry, a few times, often painfully. Particularly because we have been using AI for decades. Here is one practical example:
We used to have to hand build small terrains in video games, such as 2 kilometre by 2 kilometre levels. So you would hand paint mountains, hand paint rivers, hand place forests, sometimes even tree by tree! You would have to place villages, and cities, and it was a lot of manual work.
A small 2km by 2km terrain
Then, in the 2010s, things changed. Suddenly procedural terrain generation became much easier, due to technological advancements. You could generate 10 kilometre by 10 kilometre maps with ease! You can generate entire mountain ranges, and even volcanoes. You can generate rivers, and automatically make them intersect correctly with mountains. You can generate entire forests, and automatically make them intersect with rivers! Then you placed a city, and a few villages, and you’re ready to go. So big AAA studios did this, and gave it to playtesters.
A huge 10km by 10km terrain
And playtesters hated it. They said the worlds felt empty and repetitive. This really happened! You can ask folks at AAA studios who were there. It was a terrible, terrible time. Suddenly you had to add a whole bunch of new cities, new villages, new quests, and NPCs, to fill up this giant world, and not make it feel empty. So procedural generation didn’t remove work. It added work! It added way more work!
You had to build new towns & villages with unique environment art, NPCs with unique dialogue, both written and spoken, new quests & missions with unique stories and rewards, and all this new content had to be balanced correctly, so you also needed a new role on the team: An economy designer, for a functioning item economy.
So, procedural generation removed some work, but also added way more work. To quote someone: “Your scientists were so preoccupied with whether they could, they didn’t stop to ask themselves whether they should”. Of course said by the great Dr Ian Malcolm, played by Jeff Goldblum in Jurassic Park.
Dr Ian Malcolm, played by Jeff Goldblum, in Jurassic Park - Universal Pictures
So, creating a lot of ‘stuff’ is not helpful. Freya Holmer made a great video about this called ‘Generative AI is a Parasitic Cancer’.
A screenshot from Freya Holmer’s video
In her case she wanted to know more about the glb file format, but all the red markings you see on the results page are AI generated slop, that contain ‘stuff’, but not actually the right info about the glb file format. Much like how we generated a lot of terrain in video games, but not an interesting world.
Another topic that has come up a lot with genAI in game development is:
“What if the game automatically reacts to the player?!”
Yes, that would be cool! Which is why Left 4 Dead has an AI director, which was released in 2008. How Left 4 Dead works is that you have 4 survivors. It’s a multiplayer game with 4 players. There are zombies that they have to shoot to survive and fight through various levels. There is an AI director looking at the gameplay to ensure a fun and smooth experience. So that there are highs and lows, and a fun pacing throughout the game.
For example maybe you aren’t doing so well, so it spawns a few less zombies. But then you get your items back, and are doing well, so it spawns way more zombies. This is a well known system in the games industry, because Valve talked publicly about this. In a talk they call it the Adaptive Dramatic Pacing algorithm.
A slide from a Valve presentation
They called it a “key system of the ‘AI Director’”. This information, and much more, is in a talk called ‘The AI systems of Left 4 Dead’, by Michael Booth. Which released in 2009.
Another thing that has come up a lot with the new LLM AI wave is:
“What if NPCs could reply to anything you asked them?”
The question that you need to ask someone when this comes up is: “Do you think the reason NPCs don’t is purely the cost?”. Because it’s game design. Let’s say an NPC approaches you in a game. They say ‘Hello’. Then another NPC approaches you, and they say ‘Great weather today’. Then you walk past another NPC, and they say “My husband’s been kidnapped!”. Now which one do you think you need to talk to, to get a quest, or to advance the story? Talking to NPCs is a game design system. If you could infinitely talk to every NPC, it would get extremely confusing.
Can you generate unlimited dialogue for your NPCs? Yes, you can. We have been able to do this for a long while in text-only games.
Is that good game design? No, it hasn’t been. Games have not become more successful with infinitely talking NPCs.
Again: “Your scientists were so preoccupied with whether they could, they didn’t stop to ask themselves whether they should”
And again, I am not saying it can’t be solved! I am saying at the moment it’s not solved. So when you hear an idea like this, check up on it with a veteran from the games industry and ask: Has this been done before? And if so, why, and how? What problems did they have? What went well? Do not simply release a new version and hope it goes well, but learn from the past.
AI can be useful, but be specific
Of course there are useful applications for ‘AI’, depending on what you mean when you say ‘AI’. For example, Promethean AI released this video in 2019, long before ChatGPT, in which they have an asset system automatically tag assets correctly. And then you can ask it questions in natural language, and it shows you what assets you are looking for, without having to remember their names. It’s awesome!
Workflows like that greatly help speed up game development. They consist of natural Language and fuzzy search. Technologies which have existed for ages, and are always expanding over time.
Let me give you another example. I often use my MacBook to present at conferences and companies. So I connect my laptop to a projector or screen, and then I want to be able to set up that monitor so it displays my presentation correctly. macOS has a great feature called ‘Spotlight Search’ that allows me to search for anything on my MacBook. So, let’s search!
So I search for screen, because I want to get the share settings of my screen. But I get no good results. Then I search for monitor, because I want to adjust my external monitor. Nope, no good results. No, I have to type ‘Displays’ and then I get the options I want. Even though Screen and Monitor aren’t technically incorrect words. They make sense, they’re just not exactly correct to what the system is called. This is a great place for a natural language system to suggest me features that make more sense. I have talked about this for ages, and published another article about it in 2022.
Or what about in Blender, when I want to see the statistics of the scene, such as triangles, vertices, etc. I can hit F3, and search.
But if I type Statistics, I do not get the function I want. Instead, I have to go up in the toolbar, click a specific button, and there it is! A ‘Statistics’ function. I turn it on, and there are my statistics of the scene. An automated system would be nice to have here, that searches through all available options, and fuzzy searches their names. But, do you need an expensive LLM for this? Does it need a 2 billion dollar datacenter? Or is an algorithm fine? The return on investment (ROI) needs to be considered. There are probably cheaper and more effective ways to fix this problem, that are still considered AI, but are not considered LLMs.
So please, do take busywork out of developer hands! And please do this with automation, algorithms, and AI! It just depends on what kind of AI you mean. For example, I recently released a YouTube video about Blue Prince, and I had a script written for it. On YouTube I can just copy and paste my script into YouTube’s subtitle system, and then it automatically detects what words I say in the video, and attaches the subtitles to exactly the moments where I say those words.
It sets the subtitles up frame perfectly! It’s amazing! This saves me hours and hours of manual work! It’s awesome! It does not need to train on my voice to do this either. This technology has existed for years, long before ChatGPT came online to the public.
Ubisoft has also shown lots of interesting motion matching animation tech at SIGGRAPH in 2020, which you can also find on YouTube.
So again, game developers have used new technology to improve workflows for a long time. There are also technologies like DLSS, Deep Learning Super Sampling, that can give you higher frame rates in video games. It is awesome tech, and can run locally. DLSS was released in 2018, again, long before the ChatGPT or other LLM AI releases. So again, when you talk about AI: Be specific. When you give a talk, do a panel, host a roundtable, have a conversation with someone at work, be specific. What is ‘AI’? What kind do you mean?
There is way more to discuss on this topic, such as environmental costs, ethical & labor issues, and so much more. Please ask other folks about these! I am an expert at video game development workflows, and this talk was supposed to only be 30 minutes anyway. Please ask folks like Celia Hodent about her push for ethical AI usage, for example.
A real expert will have a historical view on technology
We are seeing the same technologies, again and again over time. We are just seeing them at different scales. For example, the money Fortnite makes from UGC is vastly larger than Team Fortress 2’s UGC items. Inherently though, we can learn from what happened before and see patterns repeat, and we can use that avoid previous mistakes. You can stand on the shoulders of giants.
So again, the takeaway here is: When you hear something fantastical, please ask a game industry veteran about it
But…
When their LinkedIn profile says that in 2016 they were a Blockchain Expert, in 2020 they were suddenly a Metaverse Expert, and in 2022 they were suddenly an AI Expert, then please considering asking someone else. Because this is someone who is moving from thing to thing, and they may not have the experience that you need to deeply understand these technologies and their history.
Thank you very much for reading this, and thank you to all the folks who provided feedback and thoughts on this presentation. I really appreciate you. Thank you to Brenden Gibbons, Arden Osthof, Christian Dumancic, Sos, Celia Hodent, and Sena. Also, apologies to Matthijs Dierckx for me still saying 'like for example' a lot in the video. It’s snuck into my vocabulary, and it’s a tautology!