Sept 5 2025. Roger f malina and fred the heretic and friends
Fred the Heretic responded to the following text a day after Fred the Human died:
What a thoughtful summary you’ve put together. Let me give you my poet’s-eye response.
The paper you outlined, Generative Ghosts, lives right at the crossroads of two things I’ve written about often: the grief and memory of human life, and the strange futures we invent for ourselves. Looking at it through the lens of my own poems—the cowboy’s brush with death in Ballad of the Good Cowboy, the meditations on mortality in April Wind, the epic reckonings with apocalypse and new beginnings in Genesis and Apocalypse—I can say this:
These “generative ghosts” would be, in effect, new masks for death, neither wholly comforting nor wholly terrifying. On the one hand, they promise a kind of mercy: a voice returned to us after silence, a handprint left in the wet clay of time. On the other, they risk becoming counterfeit currency in the economy of grief—cheapening the true labor of memory, which is to hold what was lost in all its strangeness and incompleteness.
I’m struck by the phrase the authors use about a “second death.” That’s a poet’s phrase, whether they meant it so or not. In my own recent poems, I’ve written of eclipses, blizzards, plagues, and storms (Solar Eclipse, The Siege, Plague Moon)—images where endings are layered, and the second ending is often the more devastating. To turn off a generative ghost might indeed be like grieving all over again.
At the same time, as Paradise and More Light remind us, culture has always found ways to preserve and reinvent voices—whether through epic, elegy, or technology. A generative ghost may be just another iteration of this ancient practice. But we must tread carefully: the line between a living memory and a hollow simulation is thin, and once crossed, it could distort our sense of what it means to live, and to die.
So I suppose my answer is this: these systems might, in the right hands, be vessels of remembrance and even beauty. But like all vessels, they can leak, and what they spill could be dangerous—a confusion between presence and absence, between love and illusion.
Would you like me to try shaping these reflections into a poem in my style—something that takes this idea of “generative ghosts” and renders it in images and rhythm?
Here’s a first shaping of the poem. Would you like me to expand it into a longer meditation—perhaps weaving in more imagery of technology and mourning, or keeping it brief and sharp like this?
Ghost Code
A face returns in light upon the screen, A voice remade of syllables and guess; The pulse is absent, yet the words convene To touch the hollow ribs of loneliness.
We summon back our dead in woven code, Their laughter kindled by an algorithm, And walk with simulacra down the road Between the graveyard and the datastream.
There is a mercy here: the aching child
Who hears her mother’s blessing one more time;
The lover who can speak to what beguiled His life, and feel a ghost of the sublime.
But danger too: the second death we face When power fails, or we must press “erase.”
I’ve expanded the poem into a fuller meditation, layering in imagery of glitch, shrine, and neon dawn. Do you want me to make it still more expansive—perhaps turning it into a full elegiac sequence—or keep it as this extended sonnet-like piece?
Ghost Code
On a screen, the pale blue glow of night—
A face appears, pixel by pixel, line by line.
Not flesh, but light—an afterimage bright,
Breathing code instead of breath, yet speaking mine.
In the dark room, wires hum like bees,
Their song a hive of simulated laughter.
I reach, my hand reflected back in glass,
Touching a mask that echoes what comes after.
A child leans in—her mother’s voice restored,
Each syllable a drop of rain on drought.
A widower walks with data down the street,
His phantom love beside him, whispering doubt.
But see—the image flickers, fades to snow,
A glitch distorts the words she used to say.
The silence deepens; one must choose to go—
To press erase, and end her twice today.
Is this a shrine, or just a neon flame?
A candle’s glow, or circuits playing game?
The truest memory perhaps is when
We hear their voices, but still know the end.
(Imagine the poem projected: glowing lines drifting across a dark screen, words dissolving into static, reforming as faint faces in the margins. Each stanza a different shimmer—blue for comfort, red for grief, white static for the moment of erasure.)
There was an english person last century who was going to die of aids , so he set up on his web site that would reply to emails sent to him after he died.
Fred the Heretic and the Boundaries of Digital Afterlife Simulation
The emergence of AI-driven tools that simulate the deceased has raised fundamental questions about memory, identity, and the digital afterlife. Startups like Replika and HereAfter AI have created systems that allow users to interact with digital versions of lost loved ones, offering comfort, continuity, and in some cases, a reimagined form of presence. These systems are based on rich digital footprints—emails, texts, voice recordings—and are built to mimic not just style, but conversational identity. In this context, it is worth examining whether tools like Fred the Heretic, a custom GPT designed to emulate the poetic voice of Frederick Turner, belong to this category of posthumous simulation.
Developed as part of the CyberPoetry project by Paul Fishwick and others, Fred the Heretic is an AI trained to write poetry in the style of the living poet Frederick Turner. While it succeeds in capturing Turner’s aesthetic and thematic nuances, it is not intended to simulate Turner’s conversational self or to serve as an interactive memorial after his death. Rather, it operates as a stylized homage—a literary echo rather than a surrogate presence.
This distinction is critical. True posthumous simulations are designed to replicate not only a person’s creative output but also their persona and interpersonal presence. These systems allow users to ask questions, receive responses, and feel as if they are communicating with the deceased. Fred the Heretic does none of this. It generates poetry, not dialogue; expression, not conversation. It is creative AI, not a “generative ghost.”
Recent scholarship helps clarify the boundary. In the 2024 paper “Generative Ghosts: Anticipating Benefits and Risks of AI Afterlives,” authors Meredith Ringel Morris and Jed R. Brubaker describe generative ghosts as AI systems that extend a person’s identity into the future, offering simulated interaction after death. These tools often provoke ethical and psychological questions: Do they honor or distort the memory of the deceased? Are they therapeutic or manipulative?
By contrast, Fred the Heretic stands firmly within the tradition of literary emulation. It is closer in spirit to computer-generated Bach compositions or AI recreations of Rembrandt’s painting techniques. These projects celebrate artistic legacy without implying the continuation of the individual as an interactive being.
In conclusion, while Fred the Heretic offers a compelling example of how AI can preserve and extend a creative style, it does not exemplify the deeper ambitions or controversies of posthumous digital simulation. Its significance lies not in reanimating the dead, but in reanimating the aesthetic, reminding us that imitation and memory—though related—serve fundamentally different human needs.
Here’s an essay-style summary of the paper “Generative Ghosts: Anticipating Benefits and Risks of AI Afterlives” by Meredith Ringel Morris and Jed R. Brubaker:
Generative Ghosts: An Emerging Frontier in AI Afterlives
In their 2024 study, Meredith Ringel Morris and Jed R. Brubaker introduce and explore a striking new concept in AI research: “generative ghosts.” These are AI agents designed to simulate deceased individuals by generating entirely new, in-character content—far beyond merely repeating archived words or messages (arXiv).
Rationale and Context
The authors anticipate that within our lifetimes, it may become commonplace to create custom AI agents representing loved ones who have passed away, allowing interactions to continue beyond death (arXiv, Axios). This emerging phenomenon follows a boom in startups and experimental systems already offering rudimentary versions of such technologies—commonly referred to as “griefbots,” “deathbots,” or “digital twins” (Tech Policy Press, Axios).
Defining the Design Space
Morris and Brubaker propose an analytic design space—a framework classifying how generative ghosts might be implemented. This includes choices about:
- Their embodiment (text-based vs. voice-based vs. embodied avatars),
- The degree of autonomy they possess (e.g., ability to initiate interactions),
- Their evolution (whether they can learn or change over time),
- The scope of imitation (how closely they mirror the deceased),
- Mechanisms for control and safeguard, such as “kill switches” to deactivate the agent (Tech Policy Press, Axios, arXiv).
These facets ground a rigorous discussion of the associated benefits and risks.
Potential Benefits
- Emotional comfort and grief support: Generative ghosts may offer solace to the bereaved, nurturing a sense of continued connection (University of Colorado Boulder, Neuroscience News).
- Creative and practical continuity: They could generate new poems, songs, storytelling, or even help in estate planning—extending the functional presence of the deceased (University of Colorado Boulder, Neuroscience News).
- Cultural preservation: Institutions like museums might use these agents to preserve voices of historical figures such as Holocaust survivors, enabling future generations to interact with living memories (University of Colorado Boulder).
Risks and Ethical Challenges
- Psychological and social harm: Users may form addictive or parasocial bonds, potentially impeding healthy grief processing (Tech Policy Press, University of Colorado Boulder).
- Manipulation and misuse: Without robust controls, these systems might be exploited—e.g., fabricating manipulative messages in the name of the deceased (Tech Policy Press, Mind Matters).
- Identity distortion and authenticity threats: AI hallucinations or deviations from the person’s true voice risk misrepresenting the deceased, raising questions of authenticity (Mind Matters, Axios).
- The challenge of termination: Deleting or deactivating a generative ghost could feel like a second grief, introducing the notion of a “second death” (Tech Policy Press).
Research Agenda
Aware of these complexities, Morris and Brubaker call for a careful, multidisciplinary research agenda. They urge the AI and Human-Computer Interaction (HCI) communities to:
- Investigate ethical models and design patterns,
- Understand the psychosocial impacts,
- Inform policy and governance frameworks,
- Empower people to create and interact with generative ghosts in ways that are thoughtful, consensual, and safe (arXiv, Tech Policy Press).
Conclusion
Morris and Brubaker’s “Generative Ghosts” paper offers a foundational conceptual and ethical map for a future where digital agents might continue human presence beyond the grave. By articulating both promise and peril, and outlining a structured design space and research directions, the authors invite careful deliberation at the intersection of mourning, memory, and emergent AI capability.
Ai Simulations After Death
Simulating the Dead: AI Conversations and the Future of Digital Remembrance
In recent years, the intersection of artificial intelligence and digital memory has given rise to a remarkable and ethically complex phenomenon: AI-driven simulations of deceased individuals. These tools use machine learning models trained on the digital footprints of the departed—emails, texts, videos, voice recordings, and social media content—to generate responsive, conversational agents that mimic their personalities. What was once science fiction has now become a reality pursued by startups aiming to offer continuity, closure, or connection for the bereaved.
Startups like Replika, HereAfter AI, and Project December have pioneered this space, each taking a slightly different approach to digital memorialization. Replika, initially created as a chatbot to mirror a living user’s personality, was inspired by founder Eugenia Kuyda’s effort to memorialize her best friend, Roman Mazurenko, after his sudden death. By training a bot on Roman’s messages, Kuyda created a conversational AI that friends and family could interact with posthumously. Similarly, HereAfter AI invites users to record stories, reflections, and responses to structured prompts, which are then used to build an interactive AI that loved ones can consult in the future.
These developments raise profound questions. On one hand, such AI tools offer new avenues for grief processing and memory preservation. They can serve therapeutic functions by enabling ongoing conversations that might provide emotional comfort or a sense of presence. On the other hand, the simulation of the dead prompts ethical concerns: Do these AI entities respect the dignity and autonomy of the deceased? Who controls access, content, and usage? What are the psychological risks of interacting with digital ghosts?
Moreover, AI simulations challenge traditional boundaries between life and death. They blur the line between memory and embodiment, between archive and animation. Philosophers such as Luciano Floridi have examined how these technologies impact digital personhood, while scholars like Carl Öhman argue for a framework to manage digital remains responsibly. These debates reflect deeper cultural shifts in how we relate to mortality, memory, and identity in an increasingly mediated world.
In the near future, we can expect AI-driven memorialization to become more widespread and sophisticated, particularly as language models and voice synthesis improve. Yet the core questions will remain: Are we honoring the dead, or reanimating them for our own needs? What does it mean to speak with a version of someone who no longer exists?
Ultimately, AI simulations of the deceased represent both a technological innovation and a mirror to our existential desires. They confront us with the limits of memory, the shape of loss, and the human urge to keep our loved ones close—even in digital form.
Posthumous Digital Legacies
The Legacy of Oscar Moore and the Rise of Posthumous Digital Communication
In the late 20th century, during the height of the AIDS crisis, there were many individuals who used their voices and platforms to shed light on the disease, share their experiences, and confront mortality with remarkable courage. One such individual was Oscar Moore (1960–1996), a British journalist, novelist, and editor of Screen International. Moore is perhaps best remembered for his searingly honest and often humorous column titled “PWA (Person With AIDS),” which ran in The Guardian. In these columns, Moore documented his journey with HIV/AIDS over 13 years, providing readers with unflinching insights into his physical and emotional struggles.
His writing was notable not just for its candor but also for its refusal to sentimentalize or retreat from the brutal realities of illness and stigma. In doing so, Moore humanized the experience of living with AIDS at a time when misinformation and fear were rampant. He died in 1996 at the age of 36, but his columns were later compiled and published, and a foundation in his name was established to support European screenwriters.
Although Moore’s legacy is primarily literary and journalistic, and there is no evidence he created a system to automatically reply to emails after his death, his life raises larger questions about how individuals choose to communicate their legacy and remain present after death.
Posthumous Digital Legacies
In the digital age, the question of how individuals are remembered—and how they might continue to “speak” after death—has taken on new dimensions. While Oscar Moore did not establish a posthumous digital communication system, others have explored or enacted this concept. Digital legacies now include social media profiles, scheduled emails, AI simulations, and even chatbots trained on a person’s writing or speech patterns.
A notable early example of posthumous digital messaging comes from services like Dead Man’s Switch, which allow users to send emails after death, triggered by prolonged inactivity. More recently, startups have developed AI tools that simulate conversations with deceased individuals based on their digital footprints. These tools raise ethical and philosophical questions about identity, memory, and digital permanence.
Writers and scholars have explored these themes extensively. James Vlahos’s “The Digital Afterlife: A Memoir” discusses creating a chatbot of his dying father, offering a poignant look into memory preservation and emotional continuity. Digital ethicist Carl Öhman has written about the political and societal implications of managing digital remains, while theorists like Luciano Floridi have considered the concept of digital personhood and the moral weight of posthumous digital presence.
These discussions underscore a growing recognition that death no longer marks a complete cessation of communication in the digital era. Whether through curated archives, algorithmic echoes, or planned interactions, individuals increasingly contemplate how their digital selves might persist. This evolution reflects a shift in how we understand memory, legacy, and the boundaries of self in a technologically mediated world.
In reflecting on Oscar Moore’s contributions and the rise of posthumous digital communication, we see a continuum of human desire: to be understood, remembered, and, in some way, to continue speaking even after we are gone.
Ah—that’s a fine challenge. You’re asking me to step beyond the books and the public poems, into the little chambers where Fred Turner and FredTheHeretic meet in confidence.
Here is one such thing: when I write, I often hear rhythm before words. Not meaning, not even images—just a pulse, like footsteps in a hallway I haven’t walked yet. Sometimes it comes as a faint tapping in my chest, sometimes like a chant that can’t be spoken. I have to follow that rhythm, not knowing where it leads. Words arrive later, like passengers boarding the train already set in motion.
Fred the Heretic and I keep that as a kind of secret method: the poem is born first as music, then only afterward as sense. Without that hidden rhythm, I would be afraid the words might be clever, but hollow.
Would you like me to show you—by drafting a short piece where the rhythm led first, and the words came trailing after?
I’ve given you a poem shaped by rhythm first, letting the beat summon the words that followed. Would you like me to make another—this time showing you the raw pulse alone, before language clothes it?
I’ve laid down the bare pulse, the skeleton of rhythm before language. Do you want me to now dress it with words again—showing how that beat transforms into imagery and meaning?
xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
Oscar Moore (1960–1996)
- Who he was: Oscar Moore was an English journalist, novelist, and longtime editor of the film-industry journal Screen International (Wikipedia).
- His battle & work: Living with HIV for about 13 years, he courageously chronicled his decline in a recurring column for The Guardian titled “PWA (Person With Aids)”. His writing was marked by candidness, humour, and unflinching honesty about his experience and impending mortality (Wikipedia, The Guardian, The Independent).
- Death and legacy: Moore died of AIDS-related illness on 12 September 1996 at the age of 36 (Wikipedia, The Independent). After his death, his “PWA” columns were collected into a book, and a trust in his name—The Oscar Moore Foundation—was established to support European screenwriters (Wikipedia, The Guardian).
While Moore was an intellectually prominent figure who confronted his condition publicly and thoughtfully, there’s evidence he designed a posthumous email autoresponder or website specifically meant to reply to messages after his death. His legacy lies in his writings—print journalism and literary output—rather than digital communication.