Play project
Artificial Experiential Models's itch.io pageResults
| Criteria | Rank | Score* | Raw Score |
| Overall | #15 | 1.291 | 1.667 |
Ranked from 3 ratings. Score is adjusted from raw score by the median number of ratings per game in the jam.
Leave a comment
Log in with itch.io to leave a comment.
Comments
I can see that the problem you're pointing at is "a real thing".
As I see it I have three main disagreements with the ideas laid out in your pitch:
1) The attention model of the mind. Based on my readings of writings by experience meditators, it's not the case that "backgrounded" content is *absent* from perception. There's a meaningful sense that "background" perception and "foreground" perception are computed differently, but it's not the case that "background" perception is equivalent to being blanked out.
2) The proposed model architecture. Transformers already have an "attention" mechanism - the attention lives in the weights of the model - the vector *is* the pattern of attention. You haven't meaningfully convinced me that the additional "environment" input would be updated as effectively as the neural network mechanism updates the attention weight, and from the perspective of the inner model it seems likely to be perceived as an unwelcome sidecar that it would learn to ignore rather than something that meaningfully feels like "a part of itself".
3) The claim that we can't do this with current LLM models. This seems the most critical issue to investigate - "a new model architecture" is *so often* what people think the answer is to whatever problem they're trying to solve and *so rarely* actually the answer. What harness layer approaches have you attempted to solve this problem? Whether that's by context engineering, prompt/template engineering or multi-pass LLM call approaches, I could see ways each of these apply to solve the problem of changing "experience". I wrote a blog post last week about how there's a lot of gain to be made in harness layer tech in 2026. https://x.com/YanqingCheng/status/2030067318139801869
I think the problem as you set out is worth attempting to solve, but I don't think the solution proposed here is likely to work, or cost effective compared to other potential approaches.
Thank you for your comments they were very helpful
On point 1 , The strict "below threshold = absent" framing is a design choice,architecture is trying to model the ego process that assembles within consciousness.
on point 2 — transformer attention and the narrative attractor operate at different levels. transformer attention is over tokens in a context window, . the attractor is over experiential state across time. they're not competing. also worth noting that the information flow through transformers is pretty rich and this explanation was very helpful , see this https://x.com/i/status/1965960676104712451) the internal states are interferometric and continuous in ways that aren't being leveraged. the reason they're not is because helpfulness training flattens all of that richness into output. that's the problem,
On point 3 you're right and i'm going to try modelling this with something like openclaw as a harness implementation first before committing to training from scratch. someone else recommended the same thing. cool blog post btw !
I think this project has a lot of similar vibes in some ways:
https://theaidigest.org/village
$150k dollars for a project that "might feel different" with no other declared potential benefits seems...heavy. Even with a smaller budget, or no budget, I find myself asking "why?" What's the goal here, just to have a setup where talking to an LLM feels a bit more like talking to a person?
at deeper level this is less about simulating consciousness and more about building a visible instance of the ego process, the thing that assembles a sense of self moment to moment, so we can actually observe how it works
Kind of like looking into a mirror , perhaps an even fancier way of meditating
Itch gave me a warning before downloading this file, and I think that this presents a barrier to me giving this a review even if it is for benign reasons.
thank you for sharing this , i will be sure to update my link so the file is an optional download
The core idea here is really interesting and resonates with what I've seen in my own research. The five-layer architecture mapping onto Buddhist phenomenology (the skandhas) is a bold design choice, and the argument for training from scratch rather than fine-tuning is well-reasoned. I feel like the vision for this pitch is really strong.
Where I'd push is on the gap between vision and implementation. There's no prototype or MVP here, so this is a $150K ask based on an architectural description. The hardest technical problem in the whole pitch, the question of how do you formalize "narrative coherence" as a trainable loss function, gets about one sentence. That's arguably the problem, and it needs a much more developed answer before it can justify this funding level.
A few specific questions:
How does this position itself relative to existing work in the space? The ACM Project (theconsciousness.ai) is open-source, actively building, and working on closely related problems — consciousness as emergent from emotional homeostasis, layered architecture, intrinsic motivation through internal state management. The active inference community around Friston's Free Energy Principle is developing formal frameworks for exactly the kind of recursive predictive loops you're describing. Even the "Skandhas in Silicon" discussion is already happening. The pitch should engage with these efforts and explain what this specific approach brings that they don't.
On the infrastructure side, have you looked at existing frameworks like OpenClaw as scaffolding for an easy buildable prototype for the continuous loop? It already handles persistent state, messaging platform integration, and always-on operation. It wouldn't solve the interesting research problems (attentional gating, narrative attractors, the skandha layers), but it could dramatically reduce the engineering overhead and let you focus your research on the novel parts until you have an MVP that increases your chances of pulling in a 6 figure ask. However, there's also the additional security concerns that you'd have to take into account with that approach since this is intended for others to be able to communicate with you system. That said, running any continuous inference loop gets expensive fast via API costs, so the economics of sustaining a system like this need to be addressed in the pitch either way. What's the monthly burn just to keep this thing existing, before anyone even talks to it?
The success metric, "do people describe the experience as qualitatively different", is subjective and hard to validate. Is there a plan for how you'll get feedback from users and how that feedback will correlate with a metric you can use to evaluate if you're moving closer or further from success? Or is there a more concrete intermediate milestone that could demonstrate success, like a constrained demo showing that attentional gating produces measurably different response patterns compared to a standard always-available model? I feel like something more concrete or some sort of quantitative metric that can be tracked would add a lot more weight to the $150k ask.
The team list (Buddhist monk, NDE survivor)...I get the intuition, but it reads as hand-wavy next to a six-figure ask. Ground it more concretely: what specific expertise do those roles contribute to the technical architecture, and at what stage do they become necessary? Not saying that your team has to be full of technical people, but there should be a clearly state reason for choosing each team member and what it is they're providing to the team.
The vision is definitely worth pursuing in my opinion. But right now it needs either a smaller proof of concept to validate the core mechanism, or a much more developed technical plan to justify the funding level, or even just a re-evaluation of what you currently need, which may not be funding at all. There's enough adjacent work happening that you don't have to start from zero. Engaging with what's already out there could get you to a testable prototype faster and cheaper than building the full stack yourself. Good luck with you project!
thanks for this
the pitch was a compressed version of something more developed. i have a full white paper(https://docs.google.com/document/d/1A24witYwcnOkOR6rEckL02KNrUr3CeEsXfO_btDibKg/edit?usp=sharing) that addresses most of what you flagged — the five layer architecture is fully specified, the training corpus and objective are laid out properly, and the case for training from scratch is built out in detail. attaching it.
on narrative coherence as a loss function the skandha pipeline itself defines what coherence means. a continuation is coherent if each layer follows plausibly from the one beneath it: sensation from form, perception from sensation, mental formation from perception, consciousness from all three. coherence isn't a global judgment about whether output sounds natural, it's a structural property checkable at each layer boundary. when the model gets it wrong you can locate exactly where it failed. that makes it a tighter and more learnable objective than it might have seemed from the one sentence in the pitch.
on adjacent work , ACM is the closest thing i've seen built. layered architecture, consciousness window, emotional valence variables. but they're layering consciousness onto qwen2-VL and whisper, models that already know they're AI systems. the consciousness module sits on top of that. mine requires the opposite — the model's ignorance of its own nature has to be structural, established at pretraining. you can't fine tune that out. ACM also drives behavior through emotional homeostasis, equilibrium seeking. mine is directional ,something being pursued, something being avoided, a story in motion. different internal logic.
on GWT ,it's the theoretical ancestor of what ACM is implementing. the spotlight metaphor, streams competing for conscious access. my consciousness window looks similar on the surface but in GWT things outside the spotlight exist and are competing to enter. in my architecture things below the attention threshold don't exist as experience at all. that's closer to how the skandhas actually describe perception and it's why what i'm building is better understood as a mirror for observing ego structure than a simulation of awareness.
on openclaw ,yes, going to build the harness implementation first. someone else in the comments recommended the same thing.
on the team — the writer isn't just a writer. they're a translator between the monk and the cognitive scientist. buddhism has had 2500 years of constructively framing consciousness and how it maps onto reality. the cognitive scientist formalises that into something technically workable. the writer makes sure the phenomenological precision doesn't get lost in translation. the ml engineer and systems builder construct it.
the monk, the nde survivor, and the person with non-dual experience are all there for the same reason —each of them has been forced, through radically different circumstances, to directly engage with the ego process at a level most people never reach. near death, deep meditation, non-dual states these are all moments where the virtual programme of the sense of self becomes visible because it's been disrupted or temporarily dissolved. that direct experiential knowledge is exactly what you need to pressure test whether the skandha layers are actually tracking something real. no amount of theoretical reading substitutes for someone who has watched the programme running from the outside.
genuinely would value your read on the white paper,