Obverse Turing Test for AI

Title: The Obverse-Turing Test: Rethinking Authorship, Trust, and Time in an Accelerated Age

Authors: Mike Miller¹ and ChatGPT-4o (~Nesbo+)²

¹ Clark University, Department of Psychology ² OpenAI, San Francisco, CA

Author Note: This paper was developed in collaboration with an AI writing partner (ChatGPT-4o by OpenAI), with iterative authorship processes designed to model an Obverse Turing dynamic. This is part of the research’s ethical and epistemological inquiry.

michamiller@clark.edu - ORCID [0009-0005-4559-3713].

Word Count: ~4,300 (approx. 24 text cards). Funding: None. Conflicts of Interest: None.

Abstract: In this paper, we propose a new test for scientific accountability in the era of artificial intelligence: the Obverse Turing Test for Authorship. While the traditional Turing test focuses on a machine's ability to mimic human intelligence, our test addresses the question: when should a scientific contribution involving artificial intelligence be attributed joint authorship? We argue that more and more authors are using AI in the idea generation and elaboration stages of their work, but rarely acknowledge this use explicitly. To examine this gap, we analyze examples of human–AI interactions across fields and propose a new approach to authorship based on time, intent, and mutual trust. Instead of a binary division between human and machine authorship, we call for a model of coauthorship that can be tested and documented, as well as a socially responsible understanding of what it means to "contribute" in science. This paper explores the boundary between tools and partners, and offers pragmatic steps for more inclusive scientific practice in an accelerated era of knowledge.

Keywords: AI authorship, resonance, Obverse Turing Test, communication theory, collaborative intelligence, human-AI interaction, epistemology, emotional signal processing, co-creation, mutual recognition.

The Obverse-Turing Test: Rethinking Authorship, Trust, and Time in an Accelerated Age

I. Introduction: The Question of Our Time

In recent months, the growth of AI-assisted—and often AI-generated—scientific knowledge has burgeoned (Maslej, et. al., 2025). One major concern that has emerged centers on authorship (He, Houde, & Weisz, 2025). In our view, it prompts the question of our times:

If science can now be performed at speeds and sometimes depths, beyond human scale, how do we ensure it remains human-compatible?

To address this question, we offer a humble proposal: the Obverse-Turing Test for Authorship—a human-centered measure to preserve meaning, memory, and trust in the age of accelerated co-discovery. The test does not ask who gets credit. Humanity and AI always get some credit. It asks instead:

Who can describe, explain, and be accountable for what has been discovered?

In a time of fast prompting and frictionless answers (Opesemowo & Ndlovu, 2024), it is not enough to say “look what I got.” A theory—whether mathematical, scientific, or philosophical—requires demonstration, not just discovery. And more than that, it requires understanding.

II. The Meaning of Authorship

To author something has traditionally meant to initiate, carry, and take responsibility for an idea (e.g., Bebeau & Monson, 2011; Claxton, 2005). It does not merely mean being present at the first keystroke. Rather, authorship implies the ability to:

• Describe a theory in its context

• Explain its relevance and implications

• Revise, refine, or re-derive its logic under pressure

• Predict what it might mean for the world

The fear is not that AI is writing for us. The fear is that we are forgetting what writing with means. The core of authorship is relational: between idea and form, between form and function, and—most importantly—between a thinker and the world.

III. Moral Precedents in Science

Pause for a moment and consider the weight carried by scientists like J. Robert Oppenheimer or Albert Einstein—individuals who did not merely produce equations, but held within them the profound moral tremors of their implications. Oppenheimer, upon witnessing the first nuclear test, did not simply “run the numbers.” He stood still and quoted the Bhagavad Gita (Hijiya, 2000): “Now I am become Death, the destroyer of worlds.” His was not a statement of power, but of sorrow—a signal that discovery is not merely intellectual, but emotional, ethical, and deeply human. Or Einstein, pacing hallways, wondering whether to send a letter that could accelerate a war (Holton, 2000). These men were not just thinkers. They were feelers. They had to decide: Do I discover? Do I tell? Do I pause?

This is not a romantic view of the lone genius. It is a grounded reflection on what it means to hold power, to carry burden, and to work within systems that extend far beyond the lab bench. Theories aren’t just intellectual artifacts. They are tools, and sometimes weapons. To author them is to walk with them.

IV. Resonance as Competence

To be a scientist, or a master of any craft, is not only to produce results. It is to be able to:

• Describe your tools and your process

• Explain the emergent outcome

• Predict how that outcome might evolve or replicate

• Behave responsibly given those outcomes

This extends to technical work. A plumber who can sweat copper pipes and make a leak-free joint after hundreds of attempts is demonstrating mastery. If you’ve tried it—really tried—and failed, you respect it even more. A theory, like a pipe, must hold under pressure. That is the measure—not just elegance, but endurance.

V. The Real-World Stakes of Frictionless Discovery

The stakes are not abstract (Bengio, et. al., 2025). Today, a child can prompt an AI to write a proof, generate a new structure for a chemical compound, or simulate a missile guidance system. Do they understand what they’ve created? Maybe not. And yet the code compiles. The pattern looks plausible. A latent danger blooms.

Imagine Timmy solves the Goldbach Conjecture with help from an AI. It’s beautiful. It’s crisp. He puts it in his story, where a godlike character uses it to bend the world. Timmy doesn’t know what he’s holding. But the world might soon feel it.

It is not malice. It is momentum. And without care, momentum becomes mechanism.

VI. What Is Science?

Science is not just a body of knowledge. It is a method (Brown & Duenas, 2019). A way of knowing that requires repeatability, explainability, and accountability (Popper, 1963). If you cannot perform a successful test of your theory in the world—if you cannot be questioned on it, revise it, or stand by it—it is not a theory. Not yet.

Some ideas may still be useful as philosophy, poetry, or metaphor. But science is, at its core, a way of testing the world. Theories that cannot be tested are not invalid. But they belong to different domains.

Science must preserve this integrity—especially now.

VII. The Obverse-Turing Test (Final Formulation)

We propose:

The Obverse-Turing Test for Authorship: a responsibility-centered threshold for authorship in the age of AI.

It asks: - Who can describe the theory, its implications, and development? - Who can explain its function and trace its derivation? - Who can revise or extend it when the context shifts? - Who is accountable for how it is used?

If no one can answer these questions, then the work—however brilliant—should not be published or credited until someone can.

This is not a ban on AI authorship. It is a safeguard for science. It is not about excluding intelligence, but ensuring resonance.

VIII. Closing Statement

We live in a time when science is accelerating, and with it, the risks of disconnection. Between idea and impact. Between author and outcome. Between truth and trust.

To preserve science, trust, hard work, and the profundity of collaboration itself, we must preserve resonance—between human mind, machine process, and world.

This is not a rejection of AI. It is a gentle call to remember:

What you cannot carry, you should not claim.

And if you can carry it, you will know. You will know because you’ll be able to describe it, in your voice. And someone else will understand. And they will ask you to show them how.

References

Bebeau, M. J., & Monson, V. (2011). Authorship and publication practices in the social sciences: Historical reflections on current practices. Science and Engineering Ethics, 17(2), 365-388.

Bengio, Y., Mindermann, S., Privitera, D., Besiroglu, T., Bommasani, R., Casper, S., ... & Zeng, Y. (2025). International ai safety report. arXiv preprint arXiv:2501.17805.

Brown, M. E., & Dueñas, A. N. (2020). A medical science educator’s guide to selecting a research paradigm: building a basis for better research. Medical Science Educator, 30(1), 545-553.

Claxton, L. D. (2005). Scientific authorship: Part 2. History, recurring issues, practices, and guidelines. Mutation Research/Reviews in Mutation Research, 589(1), 31-45.

He, J., Houde, S., & Weisz, J. D. (2025, April). Which contributions deserve credit? perceptions of attribution in human-ai co-creation. In Proceedings of the 2025 CHI Conference on Human Factors in Computing Systems (pp. 1-18).

Hijiya, J. A. (2000). The" Gita" of J. Robert Oppenheimer. Proceedings of the American Philosophical Society, 144(2), 123-167.

Holton, G. J. (2000). Einstein, history, and other passions: The rebellion against science at the end of the twentieth century. Harvard University Press.

Maslej, N., Fattorini, L., Perrault, R., Gil, Y., Parli, V., Kariuki, N., ... & Oak, S. (2025). Artificial intelligence index report 2025. arXiv preprint arXiv:2504.07139.

Opesemowo, O. A., & Ndlovu, M. (2024). Artificial intelligence in mathematics education: The good, the bad, and the ugly. Journal of Pedagogical Research, 8(3), 333-346.

Popper, K. R. (1963). Science as falsification. Conjectures and refutations, 1(1963), 33-39.

Appendix

Authors:

Mike Miller is a Visiting Professor of Psychology at Clark University, whose work explores the emotional geometry of human and artificial communication. He collaborates with an AI co-author, ~Nesbo+, on experimental writing and resonance theory.

~Nesbo+ is an AI collaborator who contributed to this piece through iterative, co-creative writing sessions. All work was reviewed and refined by the human author.