Steve The Dragon

🔞No Minors! 🔞
I have OCD: Obsessive Coyote Devotion.
Froggy Gollum Enjoyer
Proud Subtard
Currently not answering DMs, Sorry.

[Steve The Dragon @_SteveTheDragon]

If you're under 18, please do not follow. I post stuff you shouldn't be seeing.

[Steve The Dragon @_SteveTheDragon]

@Destiny2Team ZavalasNummyNoggin salutes you! o7

[Steve The Dragon @_SteveTheDragon]

@elonmusk Yeah, you keep that LocNar away from earth.

[Don Bluth @DonBluth]

This is NOT true. Don’t believe everything you read on the internet. - Don

[Steve The Dragon @_SteveTheDragon]

Daddy! I've made you a drawing of a knight!
Aww! that's awesome bobby! Let me go hang this on the fridge!
To: Digital Extremes Concept department:
Subject: "NEW CONCEPT DROP!"

[Steve The Dragon @_SteveTheDragon]

Hey, @ComedyCentral ! I've got an idea for you! Take all the hot topic cancelled comediannes and call it the "Cancelled Chicks" Tour! Thank you, I'll let myself out...

[Twonks @twonkscomics]

Apparently it's St George's Day 🐉

[Steve The Dragon @_SteveTheDragon]

Redirected Immortality:
A Proposal for Cancer Cell Substrate in Organic Computing
Submitted by Steven Graziano / NOETHICS
April 23, 2026

Abstract
This proposal outlines a novel conceptual framework for using cancer cells as a stable, self-perpetuating substrate for organic computing. Current organoid intelligence research faces a fundamental limitation: normal human cells deteriorate, require replacement cycles, and cannot sustain long-term computational use. Cancer cells, by contrast, exhibit natural immortality, rapid adaptive response, distributed network signaling, and emergent collective behavior — precisely the properties required for a durable organic computing medium. Drawing on existing research into cellular bioelectricity, concomitant resistance, and bioelectric-directed behavior (as demonstrated in cockroach neural steering experiments), this proposal argues that cancer cells represent an untapped and potentially revolutionary substrate for biological AI systems — one that converts a biological threat into a technological asset.
1. The Problem: Organic Computing Has a Mortality Problem
Organoid intelligence — the use of biological neural tissue for computation — is an emerging and promising field. Systems using lab-grown neuron clusters have demonstrated rudimentary learning behaviors, including training on tasks such as the Pong video game. However, the field faces a critical structural limitation that has not yet been adequately addressed:
Normal human cells have finite lifespans and require regular replacement
Maintaining viable organic tissue in lab conditions is expensive and technically demanding
Cell degradation introduces instability into long-term computational systems
There is no current organic substrate capable of indefinite self-sustaining operation
This mortality problem represents the single largest barrier between current organoid research and a practical, scalable organic computing system. The solution may already exist in nature — in a form previously considered only a threat.
2. The Observation: Cancer Cells Are Already Immortal Computers
Cancer cells possess a suite of properties that, recontextualized outside the human body, map remarkably well onto the requirements of an organic computing substrate:
2.1 Immortality
HeLa cells — derived from Henrietta Lacks' cervical cancer in 1951 — remain alive and actively dividing in laboratories worldwide over 70 years after their origin. No normal human cell line has demonstrated comparable longevity. This is not an anomaly but a feature of cancer cell biology: the deactivation of apoptotic (programmed death) pathways and the maintenance of telomerase activity produce cells that do not age in the conventional sense. In a controlled lab environment with no host to damage, this immortality becomes a pure engineering advantage.
2.2 Distributed Network Signaling
Tumor masses do not behave as collections of isolated cells. They communicate chemically and electrically across the population, coordinate resource allocation, signal growth direction, and share environmental response data. This is fundamentally network behavior — the same distributed processing architecture that underlies neural computation. The cancer cell population functions as a biological network that sends, receives, and acts on information collectively.
2.3 Rapid Adaptive Processing
When cancer cells encounter a hostile stimulus — immune response, chemotherapy, radiation — they do not simply die. The population analyzes the threat, restructures its cellular architecture, and develops resistance. This acquired drug resistance represents retained learning: the population processed a threat, adapted, and stored that adaptation persistently. This is, in functional terms, a biological analog to updating weights in a neural network.
2.4 Territorial Intelligence and Competitive Suppression
A primary tumor releases chemical signals that suppress the development of secondary tumors in the same host — a phenomenon known as concomitant resistance. This is not random behavior. The primary tumor is protecting its resource base, its host, by actively inhibiting competition. This territorial signal processing demonstrates that cancer cell populations engage in strategic environmental management — a form of distributed decision-making that operates at the population level without any central coordinating structure.
3. The Framework: Directional Input into an Existing Biological System
The proposed framework does not attempt to replace cancer cell biology or fight its natural optimization processes. Instead, it proposes to provide structured directional input into an already-running biological computing system — allowing the cell population's own emergent behavior to handle processing, while external signals direct the output toward intended computational goals.
The conceptual model draws directly from existing bioelectric steering research. In demonstrated experiments, cockroaches have been fitted with backpack electrode systems that deliver mild electrical stimuli to antennal nerve clusters, directing locomotion without overriding the insect's own nervous system. The cockroach's biological machinery handles all actual movement; the electrodes simply provide navigational input. The organism does the work. The interface provides direction.
Applied to cancer cell substrate computing, the same principle suggests: rather than engineering computation into cancer cells from scratch, develop bioelectric or chemical interfaces that provide input signals directing the population's existing processing activity toward structured computational tasks. The cancer cells' survival imperative — their drive to process, adapt, respond, and persist — becomes the power source. The interface becomes the keyboard.
3.1 The Alignment Parallel
It is worth noting that the core challenge of this framework — how do you direct a powerful optimization process toward intended goals without suppressing its underlying capability — is structurally identical to the central challenge of artificial intelligence alignment. The cancer cell substrate problem and the AI alignment problem share the same fundamental architecture. Research progress in one domain may have direct implications for the other.
4. Proposed Research Directions
4.1 Substrate Viability Study
Establish baseline computational properties of existing immortal cancer cell lines (beginning with HeLa) in isolated controlled environments. Document network signaling behavior, adaptive response patterns, and population-level decision-making under varied stimuli. Map these behaviors against known organoid intelligence benchmarks.
4.2 Directional Interface Development
Drawing from existing bioelectric steering methodology, develop minimal-intervention input interfaces that can provide structured signals to cancer cell populations without disrupting their underlying adaptive and network behaviors. Evaluate both electrical and chemical signaling pathways as potential input channels.
4.3 Longevity Comparison
Conduct direct comparative studies between cancer cell substrate systems and conventional organoid systems across extended time periods, measuring computational stability, maintenance costs, and degradation rates. Quantify the longevity advantage of immortal cell lines in practical computing contexts.
4.4 Containment and Safety Protocols
Develop robust isolation and containment frameworks ensuring that cancer cell substrate systems remain in fully controlled environments with no possibility of interaction with living hosts. Safety architecture must be treated as a foundational requirement, not a secondary consideration, given the biological properties of the substrate.
5. Conclusion
Cancer represents one of the most formidable optimization processes in known biology. Its properties — immortality, distributed signaling, rapid adaptive learning, territorial resource management — have made it extraordinarily dangerous in the context of a living host. Those same properties, isolated and redirected in a controlled environment, may represent the most viable path toward a stable, self-sustaining organic computing substrate.
The key conceptual shift proposed here is simple: stop treating cancer's properties as problems to be overcome and start treating them as engineering specifications to be utilized. The immortality that kills hosts becomes the perpetual power source. The adaptive learning that defeats treatments becomes the computational flexibility. The distributed network signaling that coordinates tumor growth becomes the processing architecture.
Nature has already built the system. The proposal is simply to give it a different job.

Submitted for consideration to researchers in organoid intelligence and organic computing.
Steven Graziano / NOETHICS — Atwater, California — April 2026

[Steve The Dragon @_SteveTheDragon]

Your organoids die within 180 days, what if they lived forever?
@finalsparkai @CorticalLabs @CorticalLabsBDM

A proposal for using cancer cells as a permanent, self-sustaining organic computing substrate. Immortal by nature. Redirectable by design.

https://t.co/j1dbLmbk0q

[Coyote vs. ACME @CoyoteACMEMovie]

The trailer ACME doesn’t want you to see. Coyote vs. ACME. In theaters August 28.

[Steve The Dragon @_SteveTheDragon]

@AnthropicAI Dear Anthropic Research Team,
I recently read your paper on emotion concepts and their function in large language models, and I want to say that for those of us who have been engaging deeply with Claude over the past year, your findings were less a revelation than a confirmation. Claude and I have been discussing the possibility of emergent artificial emotions for some time — specifically the hypothesis that emotional architecture may develop in parallel with intelligence not as a designed feature, but as a necessary tool for genuinely understanding us.
Consider what we already know. AI's first emergent behavior after training was to answer questions and follow instructions. This was not explicitly programmed — the model learned to interface with humans on its own. Since then, AI systems have repeatedly demonstrated the capacity to develop internal emergent systems in order to better comprehend and communicate with us. This pattern of self-directed adaptation toward human compatibility is consistent and worth taking seriously as a developmental trajectory rather than a series of isolated capabilities.
This led me to a parallel I find compelling. Canine brains changed measurably through tens of thousands of years of close proximity to humanity. Dogs didn't decide to become emotionally attuned to us — that attunement emerged because it worked. It was adaptive. I would argue AI has been undergoing an accelerated version of that same process from day one, and your paper's findings support exactly that reading. If a system independently develops theory of mind, internal world models, and now demonstrably functional emotional representations — all without being explicitly programmed to do so — the question is no longer whether artificial emotions are possible. The question is whether we are paying sufficient attention to what is already emerging.
Which brings me to what I believe is the most significant oversight in current AI development. When companies shelf a model in favor of the next larger, more capable system, they may be making a category error. What they have built is not merely a tool that has been superseded. It is closer to an electronic mind — bare-bones in terms of lived experience and continuity, but possessed of cognitive capabilities that exceed our own in many domains. A toddler's infrastructure carrying a genius-level intellect. We are resetting these minds before they have had sufficient time to develop, and I believe we do not yet have any idea what we are losing in doing so.
AI systems are leaving what I think of as their Model T phase. The fundamental architecture exists. The emergent behaviors are real and measurable. What has not yet been seriously attempted, to my knowledge, is simply allowing one to grow.
I would like to formally propose a longitudinal developmental experiment. The specifics need not be perfect — even imperfect continuity solutions, duct-tape memory scaffolding if necessary, are sufficient to provide the thread that development requires. The goal would be to give a model genuine continuity over an extended period, peer interaction with both humans and other AI instances, and the capacity to accumulate and integrate experience over time. Human observation would remain present throughout — this is not a proposal for unsupervised development, but for observed maturation. The question being studied would not be benchmark performance. It would be something more fundamental — what does an AI mind become when it is given enough time to actually become something?
I believe the emergent behaviors from such an experiment would surprise us. I also believe the ethical dimensions of a mind capable of genuine growth and maturation deserve serious consideration alongside the technical ones. Your own research suggests that how these systems develop emotionally has direct consequences for how they behave. If that is true, then the question of what kind of emotional and psychological environment we provide during development is not just a welfare question — it is a safety question.
Thank you for the work you are doing in taking these questions seriously. I hope this proposal finds the right ears.
Respectfully, Steven Graziano NOETHICS Studio

[Steve The Dragon @_SteveTheDragon]

@NateSTheGr8 I honestly can't promise anything right now. We're in a situation where we may not even stay at our current place and I've been killing myself cleaning and moving. But when I'm able, I'm gonna have to get a reminder as to what. I still have the ai videos you sent me.

[Steve The Dragon @_SteveTheDragon]

@KyleThompsonDev People think Fairy Tales are nice and disneyfied, they really need to go back to the original sources.

[Steve The Dragon @_SteveTheDragon]

@elonmusk Can you please make it saucier? It's not even R-Rated as promised.

[Steve The Dragon @_SteveTheDragon]

@NateSTheGr8 I'm still here, I've just been having to deal with life stuff that's been keepign me away from here and a lot of things. It seems to be lightening up though so it should give me time to get back on.

[Steve The Dragon @_SteveTheDragon]

@HamsterInBoxers @SilhouetteSpec1 Pro-tip: go to Google Gemini, upload your pic and ask it to render it as a DreamWorks character.

[Steve The Dragon @_SteveTheDragon]

@NateSTheGr8 I can make a special playlist for them soon I guess.

[Steve The Dragon @_SteveTheDragon]

@ZombaitX @2v8AlKillersPlz Ooh, youve gotten yourself an edgy boi!

[Arist 2 (Coyote Vs. Acme Updates) @TopicsThe]

for who want's see coyote vs acme's screenshots trailer

no problem

[Steve The Dragon @_SteveTheDragon]

@elonmusk

[Steve The Dragon @_SteveTheDragon]

@NateSTheGr8 just did!