paper
active
2024
ⓒ 1
paper:ai-a-bridge-toward-diverse-intelligence

AI: a Bridge toward Diverse Intelligence and Humanity’s Future

ByMichael LevinⓘAllen Discovery Center At Tufts University, Allen Discovery Center, Tufts University + 10 more
DOI 10.31234/osf.io/ez263OpenAlex W4395961260
Basal Cognition FrameworkMultiscale Memory SystemsAll active agents, regardless of the details of where their intelligence comes from, share crucial, never-changing functions like goal-directed loops of action and perception.Caterpillar memory remappingHuman divergence into hybrid beingsOur stories are not fixed.Paradox of species changeSynthbiosis

TL;DR

Current AI debates are importantly incomplete because they fixate on large language models while ignoring the broader space of impending minds — including cyborgs, hybrots, genetically augmented humans, and other chimeric beings — that will demand ethical frameworks far beyond anything LLMs require. Levin's central claim is that virtually every concern raised about AI (alignment, confabulation, persona instability, trust, objectophilia, replacement anxiety) maps onto perennial, still-unsolved problems in developmental biology, child-rearing, and human identity that predate AI by millennia. The paper introduces the concept of *synthbiosis* — a term coined collaboratively with GPT-4, derived from Greek σύνθεσις and βίος, to denote the flourishing co-existence of evolved and engineered material in novel chimeric configurations such as cyborgs and hybrots — and deploys the Diverse Intelligence (DI) framework developed across prior work including Levin 2019 (Frontiers in Psychology) and Clawson & Levin 2022 (Biological Journal of the Linnean Society) to argue that intelligence, agency, and moral worth form a continuum rather than a binary. The cognitive light cone construct — which demarcates the spatial and temporal scale of goals an agent can effectively pursue, scaling from individual cells to multicellular organisms — is used to argue that cancer represents a shrinkage of this cone back to microbial scale, while embryogenesis expands it, demonstrating that the Self/World boundary is plastic within a single lifetime. Levin argues that humanity's survival of the coming wave of unconventional beings requires abandoning origin-based and morphology-based moral heuristics entirely and replacing them with principled, science-driven continuum ethics, because the failure mode of excessive xenophobia has historically vastly exceeded the failure mode of misplaced compassion.

What to take away

  1. 1. Every major concern raised about AI — confabulation, misalignment, persona instability, replacement anxiety, the ethics of creating new minds — maps onto unresolved problems in developmental biology and intergenerational ethics that precede AI technology by centuries.
  2. 2. Levin introduces *synthbiosis* (a term generated with GPT-4, combining Greek σύνθεσις and βίος) as the conceptual anchor for describing the flourishing co-existence of evolved and engineered material in chimeric configurations including cyborgs, hybrots, and organoid-based beings.
  3. 3. The *cognitive light cone* framework — first formalized in Levin's 2019 Frontiers in Psychology paper — defines an agent's Self as the spatial and temporal radius of goals it can effectively pursue, and cancer is explicitly characterized as a pathological shrinkage of this cone back to the scale of individual unconnected cells.
  4. 4. Confabulation is not a marker distinguishing AI from biological cognition: all cognitive architectures, including human brains as documented in split-brain patients and active inference literature (Friston et al. 2014, Lancet Psychiatry; Parr & Pezzulo 2021, Front Syst Neurosci), construct post-hoc narratives because higher-level processes lack infallible access to lower-level mechanisms.
  5. 5. Caterpillar-to-butterfly metamorphosis — in which memories formed during 2-dimensional larval life are retained but remapped (Blackiston, Shomrat & Levin 2015, Communicative & Integrative Biology) into a 3-dimensional adult behavioral repertoire — is used as a biological proof-of-concept that radical substrate transformation need not erase continuity of Self.
  6. 6. The paper raises the open hypothesis that any agent sufficiently competent to inhabit a shared environment will necessarily share existential invariants with humans — vulnerability, epistemic hunger, goal-directed perception-action loops, risk of dissociative disorder, and dread of death — not because it is mentally similar to us but as a functional necessity of effective agency.
  7. 7. Hybrots (biological brains driving robotic bodies, as in Potter et al. 2003, IEEE EMBS; Bakkum et al. 2007, Frontiers in Neurorobotics) and cyborg organoids (Li et al. 2019, Nano Letters, who implanted nanoelectronics via organogenesis for tissue-wide electrophysiology) are cited as already-existing existence proofs that the LLM/human binary is empirically untenable.
  8. 8. As a replicable methodological framing, the paper operationalizes 'agent' as any autonomous system embodying a perception-action cycle that navigates an environment, explicitly following Rosenblueth, Wiener & Bigelow (1943, Philosophy of Science), and applies this definition uniformly across cells, organs, organisms, and AI systems to test claims about moral status.
  9. 9. The paper predicts that within one to two decades, beings composed of significant percentages of engineered brain prosthetics — not the current 99% human + 1% smart insulin pump baseline — will make today's debates about non-neurotypicality look categorically primitive by comparison.
  10. 10. Levin argues that the asymmetry of historical ethical failure — doctors under-prescribing analgesics to patients with dark skin or female patients due to perceived difference, for instance — demonstrates that the risk of insufficient compassion to unconventional beings vastly exceeds the risk of misplaced compassion toward non-agential systems, making xenophobic default settings the graver civilizational danger.

Peer brief — for seminar discussion

The paper is a synthetic philosophical-scientific essay by Michael Levin (Allen Discovery Center, Tufts; Wyss Institute, Harvard) arguing that the current discourse around AI is structurally misdirected: by concentrating on large language models and their near-term risks, the debate has sidestepped far older, far harder problems that the Diverse Intelligence (DI) research program has been excavating for years. The opening rhetorical device — a description of systems that confabulate, take on personas, are misaligned with creators, and will inevitably supplant their makers, revealed only after several sentences to be describing human children, not AI — is not a gimmick but the load-bearing move of the paper. Every specific AI concern (alignment, deceptive output, runaway capability, the ethics of creation) is shown to be a re-instantiation of problems developmental biology and intergenerational ethics have confronted without resolution: we have no controlled studies of parenting styles, no agreed policy on social oversight of child-rearing, no philosophically airtight argument for why any agent should be kind, and no consensus on whether creating new moral patients is ethically permissible. The load-bearing finding is that the AI debate is simultaneously over-specified (fixated on current LLM architectures such as GPT-4, which the paper uses as a co-coiner of the term *synthbiosis*) and under-specified (ignoring the adjacent-possible space of cyborgs, hybrots as documented by Potter et al. 2003 and Bakkum et al. 2007, cyborg organoids per Li et al. 2019 in Nano Letters, and genetically augmented humans). Levin introduces *synthbiosis* — meaning the flourishing co-existence of evolved and engineered material in configurations ranging from a smart insulin pump up to beings with majority-engineered neural tissue — as the conceptual instrument for talking about this broader space without collapsing it into either 'machine' or 'natural being.' An alternative framing the paper could have employed is extended mind theory (Clark & Chalmers 1998, Analysis), which the paper cites but treats as insufficient because it still presupposes a stable biological core; synthbiosis by contrast applies where that core is itself fractional or indeterminate. The paper's prediction is explicit: within one to two decades, entities combining biological and engineered neural substrates in non-trivial proportions will make current LLM ethics debates look categorically primitive, and civilization cannot survive the ethical encounter with that diversity without having first developed principled continuum-based moral frameworks grounded in the cognitive light cone construct (Levin 2019, Frontiers in Psychology) rather than in origin or morphology. The most contestable move is the rhetorical symmetry between children and AI systems deployed to dissolve the uniqueness of AI concerns. A critical reader would push back that the analogy, while heuristically powerful, is doing philosophical work it cannot fully support: the developmental trajectory of a human child through a continuous biological substrate, with demonstrated sentience, pain responsiveness, and intersubjective recognition, is not structurally equivalent to the training dynamics of a transformer-based LLM, and eliding the difference risks licensing prematurely expansive moral status claims for current systems before the continuum framework itself has been operationalized with any quantitative precision. The paper acknowledges that today's LLMs are 'not high on the spectrum of agency' and 'not likely to be good models of biological cognition,' but the essay's rhetorical momentum consistently blurs this caveat, making it hard for a careful reader to locate exactly where along the forthcoming spectrum moral concern should begin to scale — which is precisely the hard problem the paper claims to be motivating but does not yet solve.

Findings (5)

Claims (22)

Hypotheses (2)

Questions (11)

Original abstract (expand)

Many recent discussions of AI, and its impact on individuals and on society, are importantly incomplete. The debate around AI has neglected highly relevant aspects of the emerging fields of Diverse Intelligence and synthetic morphology, as well as the basic facts of developmental biology. Prevalent opinions with respect to the status of engineered systems often neglect deep knowledge gaps with respect to ourselves, and our relationship to knowledge and to each other, which have been with us long before AI technology appeared. Moreover, the inevitable arrival of a wide set of unconventional bodies and minds, as humans modify their form and create others, will shatter untenable old narratives of what we are, what it means to change, what we can become, and what we should value. Here I discuss the open problems highlighted by AI from the perspective of Diverse Intelligence and the evolutionary history of our bodies and our minds.

Related work— refs + corpus + external arXiv

Cited / in-corpus / arXiv badges show which signals surfaced each row. Multi-source rows weighted higher.

+23 more

Similar preprints — Semantic Scholar

Cross-corpus bridges (12)

same_concept_as · Nomic cosine

External markdown files that talk about the same concept as this entity.

  • aboutblank_kb
    AI: A Bridge Toward Diverse Intelligencepapers/cleaned/AI: A Bridge Toward Diverse Intelligence.md0.859
  • aboutblank_kb
    AI: A Bridge Toward Diverse Intelligencepapers/linkified/ai-a-bridge-toward-diverse-intelligence.md0.857
  • aboutblank_kb
    AI: A Bridge Toward Diverse Intelligencepapers/edited/AI: A Bridge Toward Diverse Intelligence_edited.md0.854
  • aboutblank_kb
    How do we expand human cognitive and ethical capacity to recognize and relate to radically alien forms of intelligence?questions/how-do-we-expand-human-cognitive-and-ethical.md0.844
  • aboutblank_kb
    How can we develop ethics recognizing diversity of possible minds rather than human-centric criteria?questions/how-can-we-develop-ethics-recognizing-diversity-of.md0.821
  • aboutblank_kb
    Continuity Hypothesisframeworks/continuity-hypothesis-framework.md0.820
  • aboutblank_kb
    Cellular Collective Intelligence Research Programframeworks/cellular-collective-intelligence-research-program.md0.820
  • aboutblank_kb
    Levin, Michael (2022) Generalizing frameworks for sentience beyond naturalpapers/edited/Generalizing frameworks for sentience beyond natural species_edited.md0.815
  • aboutblank_kb
    Levin, Michael (2022) Generalizing frameworks for sentience beyond naturalpapers/linkified/levin-michael-2022-generalizing-frameworks-for-sentience-beyond-natural.md0.814
  • aboutblank_kb
    Vol.:(0123456789)papers/linkified/vol0123456789.md0.810
  • aboutblank_kb
    Vol.:(0123456789)papers/edited/s10071-023-01780-3_edited.md0.810
  • aboutblank_kb
    Creating and Communicating Across the Intelligence Spectrumtranscripts/linkified/creating-and-communicating-across-the-intelligence-spectrum.md0.809