Tommy Thompson

@tommy.aiandgames.com

2387 Followers 494 Following

That Scottish AI and games guy... __ Founder of @aiandgames.com Game AI Programmer & Consultant Public speaker, educator, nuisance BAFTA Connect He-Him, PhD -- Weekly Newsletter: aiandgames.com YouTube: youtube.com/@aiandgames

The PS4 joke 'instructional video' of sharing disc-based games.
Sitting with Wes Kerr in Riots content studio
Headline about Sony and Bungie shafting their devs
Steam Game Releases by Year 2024: 185828 2025: 21439 2026: 10401
Steam AI Content Disclosed Game Releases by Review Count 2024: 2186 2025: 4703 2026: 3291
'Imagine like giant animatronic robot say. perhaps, Optimus Prime?] walking around in a theme park and entertaining guests as theyre waiting in line. Suddenly, your one hour wait ir line becomes a really fun and delightful and engaging experience, she says. "Imagine youre waiting on hold for a customer service agent playing a voice game of Trivial Pursuit. Suddenly you wouldn't mind vour 10 minute wait. You might actually be like, 'wait, I haven't finished. At its most basic, there storytelling experiences with infinite branches. where. because these experiences are new and additive, they're dynamic, thev 're personalized. You can take a story in any direction.
YouTube comment on my video on pathfinding in Alien Rogue Incursion by user soulshinobi: "Way more complicated than expected, why aren't these people just curing cancer instead."
Currently top 3 most watched videos last 48 hours on the AI and Games YouTube channel are all Alien Isolation related.
List of recently played games on Steam: Vampire Crawlers Replaced Aether & Iron Balatro The Killing Stone Dead by Daylight Raccoin STALKER: Enhanced Edition Resident Evil: Requiem
A nightmare image of fake ai generated people in front of an old man's boozer
Thumbnail for next week's video: 'Game AI's Existential Crisis (Part 1)'
Games each year with AI disclosures: 2024: 2156 2025: 4319 2026 ongoing: 1423
We would like to address questions regarding the use of AI in Crimson Desert. During development, some 2D visual props were created as part of early-stage iteration using experimental AI generative tools. These assets helped us rapidly explore tone and atmosphere in the earlier phases of production. However, our intention has always been for any such assets to be replaced, following final work and review by our art and development teams, with work that aligned with our quality standards and creative direction. Following reports from our community, we have identified that some of these assets were unintentionally included in the final release. This is not in line with our internal standards, and we take full responsibility for it. We also acknowledge that we should have clearly disclosed our use of AI. While these tools were primarily used during early production, with the expectation that these assets would be replaced prior to release, we recognize that this does not excuse the lack of transparency. We sincerely apologize for these oversights. We are currently conducting a comprehensive audit of all in-game assets and are taking steps to replace any affected content. Updated assets will be rolled out in upcoming patches. In parallel, we are reviewing and strengthening our internal processes to ensure greater transparency and consistency in how we communicate with players moving forward.
Screenshot of RE4 with caption: "Resident Evil 4 pivoted the series survival horror tropes into a more terror-laden and action focussed experience that was a critical and commercial hit." 10.6.2 Responding to Player Performance Capcom’s beloved Resident Evil franchise was transformed in 2005 courtesy of the fourth main entry in the series coming to the Nintendo GameCube. The series had become renowned for its deliberate and often obtuse orientation and design. Characters moved in an almost tank-like fashion, and players had to consider the character’s orientation relative to fixed camera points. This drove much of the...
1.1M Views 15.2K Comments 16.1K Likes 82.4K Dislikes
Grace Ashcroft in RE9
#4 The First AI-Native Title from a AAA Studio/Publisher Will Be Announced* As mentioned we’ve seen a slew of AAA titles that have used generative AI to some degree - largely with disdain and outright revulsion from the gaming masses - but there is very little that is what hype-merchants refer to as ‘AI-Native’: meaning that the gameplay experience requires generative AI technology to work. We have been hit with demo, after demo, after demo, advocating for these ideas, but of course it’s a lot of work to get from these rather limited experiences to a full gameplay. For AAA games studios in particular, you’re typically talking a 3-7 year window for any project to go from initial concept to release. So with generative AI only really blowing up in 2023, we’re still in that phase where any AAA game built with generative AI for gameplay is still in development. I think 2026 is the year we’re going to see a AAA game get announced that is AI-native.
Some studios will opt to only use commercial AI voice models either because they can’t afford voice actors, or simply don’t want to pay for them. Think small indie teams, and in the latter case mobile studios. Some studios will insist on only using real human performances as a matter of principle. Most studios will pay for voice actors for key sequences (cutscenes) and aim to have the rest achieved via TTS with trained models on a reduced fee. Balancing quality for key moments with cost-effective implementation elsewhere. This is pretty much what Embark has done with Arc Raiders. A handful of studios will opt to put more than 50% of their VO budget towards use of celebrities such as sports professionals, and actors from film and television, and have the rest of the voice work be TTS implementations by trained voice actors. i.e. Remember when Bethesda splashed all their VO budget on Patrick Stewart for the first 5 minutes of Elder Scrolls: Oblivion? Imagine that, but a lot worse.
Perhaps most encouraging is the House of Lords reports in their own consultations that many are in agreement that while we should seek to have transparency disclosures similar to the EU AI Act, many felt they’re don’t go far enough - a point I largely agree with. There is a need for even more granularity on how copyrighted works are adopted as part of training. This leads to a broader discussion of the viability of it all, and also highlighting the inappropriate behaviour of big tech. While some bodies ask for a virtual line-by-line breakdown of every URL visited - which is arguably quite easy to produce - Google argued that “you are asking us to summarise the internet” - which speaks to the level of theft that has occurred in training their generative models. Meanwhile OpenAI expressed concerns over how this would expose “[AI] model developers’ trade secrets”.
In summary, it’s pretty damning. The House of Lords calls out the previously advertised plans of the government as means to “drift towards tacit acceptance of large-scale, unlicensed use of creative content”. With any gains being made from this arrangement being given “to a small number of US-based firms while harms to UK creators grow.” Yeah, they did not mince their words. The house has made six suggestions on where the government should go next. Amongst those we find the following: Setting out clear guidance on AI and copyright that will remove rights for companies scrape data in an opt-out model. Set out protections both for ‘deepfakes’ and digital clones, but also necessary transparency obligations for training AI models. Establish a market for licensing UK IP for AI companies should they wish to use it, and work to prioritise development of UK-based AI models.
Since the consultation closed in February of last year, the government has been working to push forward with its initiatives, albeit with a bit of a reset given the overwhelming consensus on AI licensing, but also a change of minister in charge. Having spent just over a year in office Peter Kyle MP moved from his position as Secretary of State for the Department of Science, Innovation and Technology (DSIT), and moved into a new role as Business Secretary. It was news I welcomed at the time, given Kyle had shown zero effort to actually engage with the creative industries, yet had numerous meetings with virtually every major player in AI. Perhaps unsurprising that his opinions seemed a little skewed towards screwing over IP holders in pursuit of sunny uplands.
I am on two block lists: one for "AI Haters", the other "Trash to Block", because I write about AI.
modern PC games at a time when most were still developed for MS-DOS. Released in August of 1996 - confusing I know - the port was the last project Newell worked on at Microsoft, leaving but four days alongside colleague Mike Harrington to found their own game development studio called Valve. Throughout the early to mid-90’s PC gaming was expanding into more adventurous and mechanically rich genres. Space combat flourished with Wing Commander sequels and LucasArts own take on the genre through the Star Wars franchise - with 1993’s X-Wing and 1994’s TIE Fighter 15. Meanwhile first-person shooters began to dominate the PC space; Raven Software’s Heretic (1994) and Hexen: Beyond Heretic (1995) - shown in Figure 10.10 - were built on id Softare’s DOOM engine pivoted from science fiction and demons to dark medieval fantasy and well, demons. Meanwhile other companies were cashing in on the DOOM craze, ranging from the sublime Rise of the Triad (Apogee Software16, 1995) to the ridiculous William Shatner’s Tekwar (Capstone Software, 1995). But arguably the third major ‘pillar’ of the PC community, and one that is very important in the space of Game AI, was in real-time strategy (RTS) games
Cover of the book
Photo of my copy of Super Nintendo: How one Japanese company helped the world have fun.
Front cover of 'Play Nice: The Rise, Fall, and Future of Blizzard Entertainment'
Tis me. Wearing sunglasses. At the train station.
SteamDB info showing AI content disclosures: 2023: 133 2024: 2148 2025: 4313
YouTube Shorts: You now have access to a bundle of new generative AI tools that make creating shorts easier and more fun.
A series of YouTube videos that are flagged in the AI and Games content id (copyright flagging) backend.
Our subscription tiers on our website aiandgames.com, with annual individual subs at £30 / $40 a year.
In summary, I feel that the sad truth of the situation is that we now exist in an age where so much emphasis is placed on the consumer understanding how and why AI technology is being employed in commercial products. By understanding this, you can make a decision that reflects your own personal politics. I have condensed everything from this piece into the questions below, and now I ask you - the reader- to consider this in the context of any of the AI-laden games mentioned in this issue: Livelihoods: Does the use of (machine learning/generative) AI impact the livelihoods of workers involved in the project? Replacement: Do you feel the use of AI in the project is a means to replace professional talent as a whole? Enhancement: Does the use of AI fail to enhance the quality of the final product? Substitution: Do you think the features in this game can be achieved at similar or acceptable levels of quality without generative AI? If your answer is yes to any of these questions, then you have to ask yourself whether you are willing to part with your hard-earned cash. It’s sadly just another level of diligence required if you’re aiming to try and engage in any form of (near) ethical capitalism. Sadly adding to the 101 other political issues that might influence your purchase habits.
The Arc Raiders situation is undoubtedly complicated, but it speaks to broader issues with the ethical adoption of AI technologies in game development. Perhaps critically, I think it raises the importance of consumers being conscious of how these technologies are being adopted, and the impact it has on the livelihoods of others, in a way that we historically have not had to think about with AI in game development until the last couple of years. From my perspective, the key thing we should be thinking about is as follows: When we utilise AI technology for every gain achieved, who is losing out as a result? In this context, I think voice actors, players, and the industry at large are losing out from this decision. Let’s dig into why I think that is.
The phrase "I'm Tommy Thompson, and this is AI and Games" in the style of the Zohran Mamdani campaign font.
Yetunde Longe-Folajimi presenting on explainable metrics on pathfinding for level generation.
Organisational team photo
Instead, the loudest of AI hype merchants try to reframe game AI as the things generative AI is more likely to address successfully. It then leads to us talking about the same things over and over again… Non-player characters that can utter realistic and emergent dialogue. Of running emergent narrative systems that can generate infinite stories. Of crafting NPCs that can react to human players through natural language processing. This of course doesn’t cover the full gamut of ideas being raised - given much of the conversation is on cost cutting and changing existing paradigms - but let me stress for right now I’m focussing purely on game AI here. Now note that none of these things are topics I’ve raised thus far in this piece, and that’s because while they are a challenge, they often don’t really fall under what game AI tackles.
So to get into the meat of my thesis, I want to talk first about the immediate future of game AI, and some of the concerns that have rattled around in my head the past few years. These largely fall under four specific themes: Game AI techniques as they stand are struggling under the weight of contemporary game design. There is a need for ongoing experimentation in game AI methods, but game productions seldom have the time to explore new ideas. As a result, we confine our thinking to designing against known systems, rather than design systems for purpose. Amongst all of this, generative AI is often incorrectly held aloft as a solution, failing to understand what the real problems are, and what benefits machine learning could bring to this sector.
This is Not a Rant About Generative AI, but it is one of the factors that has led to my writing this piece. Rather this is a sentiment that has been brewing in my mind for the past couple of years, and I figured it was time I put it all down in writing, and gave it a voice. By making my thoughts on this a little clearer, I want to drive a discussion in our community over how we address it. This Thesis is a Reflection of the Fragility of the Video Games Industry. I think a lot about how the games industry is this big, expansive, and powerful sector, but it is equally one that is often fraught with challenges, and relies on inconsistent and unreliable strategies in how it funds projects, how it nurtures talent, and how it builds technology.
Me holding 'Crit' in my hand. They look cute as fuck.
Art of Absolum and Shinobi
'Playing Now' collection on Steam shows: Alien: Rogue Incursion - Evolved Edition Aliens: Dark Descent Amnesia: The Bunker Assassin's Creed: Shadows
Henry Cavill / Jason Mamoa meme where the latter sneaks up on the former. Cavill text: Tommy (me) focussing on his backlog. Mamoa text: The Alien: Isolation sequel
Robin Baumgarten's A* Mario bot from 2009
A Thursday Murder Club themed photo board
Unreal Deal Left 4 Dead Bundle Just Cause 2 Lucasarts Jedi Knight Bundle All in 2012