Publications Archives - Matthew E. Gladden: Research and Publications

Archive by category "Publications"

GPT models’ learning and disclosure of personal data: An experimental vulnerability analysis

Medium.com • April 10, 2023

SUMMARY: The possible gathering, retention, and later dissemination of individuals’ personal data by AI systems utilizing Generative Pretrained Transformers (GPTs) is an area that’s of growing concern from legal, ethical, and business perspectives. To develop a better understanding of at least one aspect of the privacy risks involved with the rapidly expanding use of GPT-type systems and other large language models (LLMs) by the public, we conducted an experimental analysis in which we prepared a series of GPT models that were fine-tuned on a Wikipedia text corpus into which we had purposefully inserted personal data for hundreds of imaginary persons. (We refer to these as “GPT Personal Data Vulnerability Simulator” or “GPT-PDVS” models.) We then used customized input sequences (or prompts) to seek information about these individuals, in an attempt to ascertain how much of their personal data a model had absorbed and to what extent it was able to output that information without confusing or distorting it. The results of our analysis are described in this article. They suggest that – at least with regard to the class of models tested – it’s unlikely for personal data to be “inadvertently” learned by a model during its fine-tuning process in a way that makes the data available for extraction by system users, without a concentrated effort on the part of the model’s developers. Nevertheless, the development of ever more powerful models – and the existence of other avenues by which models might possibly absorb individuals’ personal data – means that the findings of this analysis are better taken as guideposts for further scrutiny of GPT-type models than as definitive answers regarding any potential InfoSec vulnerabilities inherent in such LLMs.

Medium.com • March 12, 2023

SUMMARY: The crafting of optimal input sequences (or “prompts”) for large language models is an art and a science. In this article, we conduct an exploratory analysis of 4,080 sentence-completion responses generated by ManaGPT-1020. This model is an LLM that has been fine-tuned on a corpus of scholarly and popular works from the domain of management and organizational foresight, with the aim of engineering a model that can produce texts containing novel insights into the emerging impact of advanced AI, social robotics, virtual reality, and other “posthumanizing” technologies on the structure of organizations and our human experience of organizational life. More particularly, we investigate how the length and quality of texts generated by the model vary in relation to “modal hints” that are supplied by a user’s input sequences. Such hints take the form of modal verbs and phrases that suggest the degree of possibility, probability, or logical or moral necessity that a completed sentence should reflect. Our preliminary analysis suggests that such “modal shading” of prompts can have at least as great an impact on the nature of the generated sentences as the identity of the subject that a user has chosen for a given sentence.

ManaGPT: Generated text length by input-sequence subject

“ManaGPT: 4,080 NLP prompts and generated texts. Output from an LLM trained on a corpus from organizational futures studies” • March 26, 2023

SUMMARY. This dataset includes 4,080 texts that were generated by the ManaGPT-1020 large language model, in response to particular input sequences. ManaGPT-1020 is a free, open-source model available for download and use via Hugging Face’s “transformers” Python package. The model has been trained to generate analysis, predictions, and recommendations regarding the emerging role of advanced AI, social robotics, ubiquitous computing, virtual reality, neurocybernetic augmentation, and other “posthumanizing” technologies in organizational life.

ManaGPT-1020 • March 23, 2023

SUMMARY: ManaGPT-1020 is a free, open-source model available for download and use via Hugging Face’s “transformers” Python package. The model is a 1.5-billion-parameter LLM that’s capable of generating text in order to complete a sentence whose first words have been provided via a user-supplied input sequence. The model represents an elaboration of GPT-2 that has been fine-tuned (using Python and TensorFlow) on a specialized English-language corpus of over 509,000 words from the domain of organizational futures studies. In particular, the model has been trained to generate analysis, predictions, and recommendations regarding the emerging role of advanced AI, social robotics, ubiquitous computing, virtual reality, neurocybernetic augmentation, and other “posthumanizing” technologies in organizational life.

“8-Bit Mystique: An Ingardenian Aesthetic Analysis of the Appeal of Retro Computer Games”

In Roman Ingarden’s Aesthetics and Ontology, edited by Leszek Sosnowski and Natalia Anna Michna • London: Bloomsbury, 2023

ABSTRACT: Recent years have seen revived interest in 8-bit computer games developed in the 1980s and the increasing popularity of “8-bit-style” or “retro” games designed to imitate their look. Judged objectively, 8-bit games appear far more “primitive” than typical contemporary video games, leading some to suggest that they are deficient works of art whose resurgent popularity results solely from nostalgia. However, by drawing on Ingarden’s analysis of artworks as schematic constructs, we argue that 8-bit-style games’ “primitiveness” is actually a form of indeterminacy that can generate singularly meaningful aesthetic experiences by allowing (and requiring) players to perform a uniquely enjoyable kind of concretization that is impossible with more “sophisticated” contemporary games. We also draw on Ingarden’s account of the “life cycle” of a work of art to show how 8-bit games’ pattern of initial popularity, neglect, and revival reflects an organic vitality demonstrated not by kitsch but by exceptional artwork.

A joint range model created with Comport_AI

Medium.com • March 12, 2023

SUMMARY: This text is Part 3 of a three-article series on “Advanced modelling of workers’ future performance ranges through ANNs with custom loss functions.” It demonstrates how custom ceiling and floor models can be combined to create a composite prediction interval that can outperform simpler models based on MAE or SD in forecasting the probable range of workers’ future job performance.

Medium.com • March 12, 2023

SUMMARY: This text is Part 2 of a three-article series on “Advanced modelling of workers’ future performance ranges through ANNs with custom loss functions.” It investigates the mechanics of independently modelling the likely ceiling and floor of the range of a worker’s probable future job performance using separate artificial neural networks with custom loss functions.

Medium.com • March 12, 2023

SUMMARY: This text is Part 1 of a three-article series on “Advanced modelling of workers’ future performance ranges through ANNs with custom loss functions.” It explores why it’s useful to predict the probable ceiling and floor for an employee’s future performance – and why it’s difficult to do so effectively, using conventional methods based on mean absolute error or standard deviation.

Comport_AI™ (version 0.3.22) • March 5, 2023

ABSTRACT: Comport_AI is a free open-source HR predictive analytics tool in the form of a Python-based web app that uses advanced machine learning to forecast the likely range of a worker’s future job performance. Rather than mechanistically deriving the predicted ceiling and floor of a worker’s future performance from a single predicted target value using calculations based on MAE or SD, Comport_AI treats the likely ceiling and likely floor of a worker’s performance during a future timeframe as independent entities, which are modelled by artificial neural networks whose custom loss functions enable them to formulate prediction intervals that are as small as possible, while being just large enough to contain a worker’s actual future performance value, in the vast majority of cases. This allows more precise, nuanced, and useful forecasting of workers’ future job performance. Comport_AI utilizes TensorFlow, Keras, scikit-learn, FastAPI, Uvicorn, Jinja2, NumPy, Pandas, and Matplotlib. It’s developed by Matthew E. Gladden (with support from Cognitive Firewall LLC and NeuraXenetica LLC) and is made available for use under GNU General Public License Version 3.

“Factory Workers’ Daily Performance & Attrition: Data with rich causal relationships, for testing machine-learning approaches” • July 23, 2022

SUMMARY. This synthetic dataset was produced with version 0.3.15 of Synaptans WorkforceSim for distribution on Kaggle. It contains 18 months’ worth of daily performance and attrition data (411,948 observations) for a factory whose organizational structure comprises 508 workers. Due to employee turnover, a total of 687 persons appear in the dataset. The dataset’s observations cover both regular daily events (like workers’ attendance and daily level of Efficacy) and special one-time events (like accidents, an employee’s termination, or the onboarding of a new employee). A unique feature of the dataset is diverse causal relationships “hidden” within the data that are waiting to be uncovered through machine learning.

Synaptans WorkforceSim™ (version 0.1.081) • December 31, 2021

ABSTRACT: This software has been developed for educational and research-related purposes by NeuraXenetica LLC and is distributed free of charge. It creates a virtual population of simulated factory employees with diverse personal characteristics and abilities and then simulates their daily workplace behaviors and job performance through a specified number of days. Users are able to adjust the values of arguments including the number of persons employed in the factory; the number of days of activity to simulate; the minimum and maximum age of workers; and the mean value and standard deviation for employees’ characteristics such as their attendance rate, industriousness, eagerness and ability to learn, ability to teach others, degree of conscientiousness, ability to inspire and lead others, positivity of attitude, and degree of supportiveness displayed toward their fellow workers. Once such arguments have been specified, the software generates a quasi-randomized virtual workforce possessing such characteristics; simulates the activity of the virtual workforce during the desired period; and then presents the results through a series of visualizations. By building such a model of a simulated factory workforce, it becomes possible to manipulate variables, test hypotheses, and explore causal relationships in a way that isn’t feasible or desirable with real human beings in a real-world factory setting. For example, the software can facilitate investigation of the kinds of organizational elements, psychological factors, and interpersonal dynamics that have an influence on the motivation, satisfaction, and job performance of employees in a factory setting. A user can also examine the manner in which changes to a virtual workforce’s size and to the psychological characteristics of individual employees impact the structure and performance of larger organizational units such as production teams and entire shifts. The software also makes it possible, for example, to investigate the extent to which variations in the performance of different managers’ teams or shifts may result from random differences in their subordinates’ personal characteristics, rather than from differences in the managers’ own managerial abilities.

Revisiting Ingarden’s Theoretical Biological Account of the Literary Work of Art: Is the Computer Game an ‘Organism’?

Horizon: Studies in Phenomenology 9, no. 2 (2020)

ABSTRACT: From his earliest published writings to his last, Roman Ingarden displayed an interest in theoretical biology and its efforts to clarify what distinguishes living organisms from other types of entities. However , many of his explorations of such issues are easily overlooked, because they don’t appear in works that are primarily ontological, metaphysical, or anthropological in nature but are “hidden” within his works on literary aesthetics, where Ingarden sought to define the nature of living organisms in order to compare literary works to such entities. This article undertakes a historical textual analysis that traces the evolution of Ingarden’s thought regarding the nature of the literary work of art as an organism-like entity and uncovers its links with the simultaneous development of his systems theory and its central concept of the “relatively isolated system”: for Ingarden, a literary work and an organism are each a systematically transforming, “living, ” functional-structural whole that comprises a system of hierarchically arranged and partially isolated (yet interdependent) elements whose harmonious interaction allows the literary work or organism to fulfill its chief function. Having completed that historical analysis, we test Ingarden’s assessment of works of art as organism-like entities in a novel context by investigating the organism-like qualities of the contemporary computer game; insofar as their AI-driven behavior displays a form of agency, such games might appear to be even more “alive” than traditional works of art. We show that Ingarden’s conceptual framework provides a useful tool for understanding the “organicity” of such games as works of art, despite the fact that they differ qualitatively from those art forms with which Ingarden was directly familiar.

Ingarden’s Systems-Theoretical Philosophical Anthropology as a Tool for Analyzing the Technological Posthumanization of Persons and Societies (M. Gladden, 2020)

In In Roman Ingarden and His Times, edited by Dominika Czakon, Natalia Anna Michna, and Leszek Sosnowski, pp. 109-26 • Kraków: Księgarnia Akademicka, 2020

ABSTRACT: While Roman Ingarden’s ontology and aesthetics have been widely studied, relatively little attention has been paid to his philosophical anthropology – despite the central role that it plays within his thought. Here we draw on the concept of the “relatively isolated system,” developed by Ingarden over more than three decades, in order to show how his philosophical model of the human being as a three-layered emergent whole can be understood as a particular application of his more generalized systems theory. Having reconstructed Ingarden’s systems-theoretical philosophical anthropology, it is argued that it provides a uniquely valuable methodological approach and tool for investigating those emerging processes of technological posthumanization that are diversifying and transforming human societies by expanding them to incorporate new types of non-human intelligent social actors (e.g., increasingly sophisticated social robots and AI) and “otherly” human beings (e.g., individuals whose capacities have been altered through neuroprosthetic augmentation). Conventional philosophical investigations that take as their starting point the status of human beings as biological, intentional, or moral beings often focus on the ways in which contemporary social robots and AI lack such status and thereby differ radically from human beings. However, by starting from the fact that all such entities are manifestations of relatively isolated systems, an Ingardenian systems-theoretical philosophical anthropology can highlight previously unappreciated similarities shared by the “naturally” human, otherly human, and non-human intelligent social beings expected to coexist within increasingly posthumanized societies.

Utopian Confederation: Conspectus (front cover)

Volume 01 in the Utopian Confederation Sourcebook series • ISBN 978-1-944373-83-2 • Mnemoclave, 2020 • 62 pages

SUMMARY: This book offers an introduction to the Utopian Confederation RPG series and the peaceful, prosperous, and high-tech future society that provides the setting for its adventures. In this world, the island-republic of Utopia isn’t an imaginary land; it’s a diplomatic, economic, techno-logical, and cultural trailblazer that has succeeded in unifying the world’s nations under a ban-ner of peaceful collaboration – thanks largely to the Utopian mindset that combines a strong rationality and pursuit of scientific knowledge with a social and political philosophy that’s grounded in a deep spirituality and theological sensitivity.

Three Axes of a VR-Facilitated Virtual World

AVANT: Trends in Interdisciplinary Studies 11, no. 2 (2020)

ABSTRACT: In this text it is argued that immersion in virtual reality (VR) with the aid of contemporary VR equipment may offer access to novel types of virtual worlds that differ qualitatively from the “real” world and from other types of fictional worlds. The text begins by (a) distinguishing between VR systems, virtual environments, and virtual worlds; (b) showing how the virtual worlds facilitated by VR systems resemble and differ from the “virtual worlds” created in one’s mind when, for example, reading a novel or watching a film; and (c) identifying necessary and optional elements of a VR-facilitated virtual world. Employing a phenomenological approach that draws on the thought of Ingarden and Norberg-Schulz, it is shown that a visitor to a VR-facilitated virtual world can (and frequently does) shift his or her conscious attention along three different “axes”. First, one’s attention can move “horizontally” between the media that disclose the virtual world through different senses. Second, one’s attention can shift “vertically” between the virtual world’s different ontological strata, including its layers of myriad atomic stimuli; distinguishable elements that possess spatiotemporal extension; assemblages of elements that have a context and relations but lack individual meaning; glimpses that build up a lattice of meaning and contribute to one’s knowledge of the world; and the virtual world envisioned as a coherent mentally concretized whole. Third, one’s attention can shift “interspatially” between the many different overlapping constituent spaces of the virtual world, including its perceptual, concrete, natural, built, identifiable, technological, emotional, social, economic, political, cultural, ecological, and possibility spaces. This triaxial phenomenological framework can shed new light on the rich and diverse ways in which VR-facilitated virtual worlds manifest themselves as emergent wholes constituted within human consciousness; also, it suggests approaches by which visitors might more proactively mentally explore and come to inhabit such virtual worlds.

Phenomenology of the Gameworld (front cover)

ISBN 978-1-944373-74-0 • Defragmenter Media, 2019 • 362 pages

Good video games transport us to incredible new worlds. We don’t just enter such unforgettable “gameworlds” when we play first-person RPGs with high-resolution graphics; even relatively simple 2D puzzle or strategy games with retro visuals can immerse players in worlds that are beautiful, terrifying, mysterious, or moving, that are brutally realistic or delightfully whimsical.

The process by which a particular gameworld emerges is a symbiosis between developer and player: the game system presents a carefully architected stream of polygons and pixels, which somehow leads the player’s mind to construct and explore an intricate world full of places, people, relationships, dilemmas, and quests that transcend what’s actually appearing onscreen.

This book is meant for developers who want to create games that will evoke richer and more memorable gameworlds in the minds of their players. Drawing on insights from ontology and philosophical aesthetics, it provides you with conceptual frameworks and concrete tools that will enhance your ability to design games whose iconic gameworlds encourage the types of gameplay experiences you want to offer your players.

Once you’ve undertaken this philosophical and artistic journey, you’ll never look at your games – or their gameworlds – in quite the same way again.

In Proceedings of the European Society for Aesthetics: Volume 11, edited by Connell Vaughan and Iris Vidmar Jovanović • European Society for Aesthetics, 2019

ABSTRACT: This work draws on Ingarden’s systems theory to develop a phenomenological aesthetic account of the kinds of reason-defying buildings that cannot exist as physical structures in the real world but which are frequently encountered within the virtual gameworlds of computer games. Such “impossible” buildings might, for example, take the form of colossal biological entities or violate established principles of physics or geometry. First, the evolution of Ingarden’s systems theory is traced, and an account of his mature systems theory is presented: pivotal is his concept of the “relatively isolated system” whose contents are partially engaged with and partially sheltered from the external environment via the system’s complex array of semipermeable boundaries. By applying Ingarden’s thought in a novel way, a systems-theoretical phenomenological architectural aesthetics is then formulated that conceptualizes the “building” as a set of overlapping physical, informational, and psychosocial boundaries that generate interior spaces that possess rich structures and dynamics and mediate their occupants’ relationships with the world. Using this conceptual framework, it is shown how the systems-theoretical properties of real-world buildings and virtual gameworld buildings can (and often do) radically differ. Three types of “impossible” gameworld buildings are analyzed: (1) the floating castle that is a recurring element of fantasy games; (2) the shapeshifting haunted mansion that appears not infrequently in horror games; and (3) the high-tech facility that functions as the gigantic “body” of an AI, which is common in science-fiction-themed games. This aesthetic framework may be of value to game developers seeking to employ techniques of “hyperdeconstruction,” “hyperfolding,” or architectural posthumanization to design more memorable and meaningful gameworlds.

Twarde światło i architektura kosmiczna: Analiza fenomenologiczno-estetyczna filmu Tron jako wzorca paradygmatu ‘Siatki Cyberprzestrzennej’ (M. Gladden, 2019)

In Rejestry Kultury (Perspektywy Ponowoczesności, vol. 9), edited by Ksenia Olkusz, pp. 481-501 • Wrocław: Facta Ficta Research Centre, 2019

ABSTRACT: W niniejszym rozdziale przedstawiona zostanie fenomenologiczno-estetyczna analiza sposobu, w jaki użycie namacalnego, „ucieleśnionego” światła jako budulca architektonicznego w filmie Disneya Tron z 1982 roku przyczyniło się do założenia paradygmatu konstrukcji i wizualizacji immersyjnych światów elektronicznych. Paradygmat ten (Siatka Cyberprzestrzenna) pozostaje aktualny w dzisiejszej erze rosnącej popularności technologii rzeczywistości wirtualnej. Analiza obejmuje dwie części. Po pierwsze, rozważane będzie użycie „twardego światła” jako elementu architektonicznego w świecie elektronicznym, przedstawionym w filmie Tron. Druga część analizy obejmie sposób, w jaki użycie twardego światła w Tronie przyczyniło się do stworzenia jednego z podstawowych paradygmatów albo metafor konstruowania cyberprzestrzeni i wizualizacji środowisk wirtualnych, to jest, paradygmatu Siatki Cyberprzestrzennej.

International Journal of Research Studies in Management (2019)

ABSTRACT: The growing use of advanced AI, ambient intelligence, augmented reality (AR), and virtual reality (VR) technologies of the sort found within the emerging cyber-physical smart workplace has been described as enabling new forms of human-computer interaction (HCI) that are “magical” in nature. This study shows that from an anthropological perspective, such a workplace environment can indeed be understood as “magical”; however, that “magicality” is a double-edged sword that can potentially both enhance and damage user experience (UX) for workers and other occupants of such environments. First, by analyzing existing social anthropological and philosophical anthropological accounts of magic, typical elements of magical practice are identified. Using Nielsen’s empirical analysis of HCI usability heuristics as a basis, a prospective heuristic evaluation is then carried out for the usability of a generic “magical” environment, in order to identify elements of magical practice that might be expected to enhance or impair user experience when they are required for interaction with the environment. A more specific heuristic usability evaluation is then performed for the “magical” aspects of HCI created by two kinds of constituent technologies that are typical for a cyber-physical smart workplace: those of (a) ambient intelligence and IoT-enabled systems and (b) AR and VR systems. It is shown that the magical aspects of HCI within the emerging cyber-physical smart workplace differ significantly in their potential UX impacts from the magicality involved with earlier forms of computing, and the implications of this fact for the management of future workplaces are identified and discussed.

ISBN 978-1-944373-30-6 • Defragmenter Media, 2019 • 322 pages

The image of the “megacorp” – the ruthless, sinister, high-tech global conglomerate that’s grown so large and powerful that it has acquired the characteristics of a sovereign state – is one of the iconic elements of cyberpunk fiction. Such a megacorp maintains its own army, creates its own laws and currency, grants citizenship to employees and customers, and governs vast swaths of cyberspace and the physical world. If it allows traditional governments to survive in some vestigial form, it’s only so they can handle those mundane tasks that the megacorp doesn’t want to deal with itself. By these standards, contemporary companies like Amazon, Apple, Facebook, Google, Microsoft, ExxonMobil, and Walmart aren’t (yet) “megacorps”; they’re the playthings that megacorps gobble up to use for spare parts.

This volume develops a comprehensive intellectual history of the megacorp. It locates forebears of the cyberpunk megacorp not only in earlier fictional works like Čapek’s R.U.R. (1921) and Von Harbou’s Metropolis (1925) but in a string of real-world organizations ranging from the 17th-Century British and Dutch East India Companies to the Pullman Palace Car Company, the Ford Motor Company, and late 20th-Century Japanese keiretsu and South Korean chaebol – as well as in the nearly indestructible oligopolistic “megacorp” described in the pioneering theory of American economist Alfred Eichner.

By investigating the nature of the cyberpunk megacorp as a political entity, commercial entity, producer and exploiter of futuristic technologies, and generator or manipulator of culture, differences are highlighted between the megacorps of “classical” cyberpunk and post-cyberpunk fiction. Classical cyberpunk megacorps – portrayed in novels like Gibson’s Sprawl trilogy, films like RoboCop and Johnny Mnemonic, and games like Cyberpunk, Cyberspace, and Syndicate – are often ostentatiously malevolent and obsessed with short-term financial profits to the exclusion of all else; the over-the-top depictions of such companies serve a dramatic purpose and are not offered by their authors as serious futurological studies. On the other hand, the more nuanced and philosophically rich portrayals of megacorps in post-cyberpunk works like Shirow’s manga The Ghost in the Shell reveal companies that are less overtly evil, possess a broader and more plausible range of long-term strategic goals, and coexist alongside conventional governments in a state of (begrudging) mutual respect. Yet other works like the game Shadowrun depict companies that combine elements of both classical cyberpunk and post-cyberpunk megacorps.

Drawing on such analyses, the volume concludes by exploring how the idea of the post-cyberpunk megacorp anticipated a new type of real-world megacorp – the unfathomably large, fast, and complex “rhizocorp” – that’s now being made possible through ongoing revolutions in the exploitation of robotics, AI, and the Internet of Things – and which threatens to become the dominant economic, political, and sociocultural power of our technologically posthumanized future world.

Category: Publications

GPT models’ learning and disclosure of personal data: An experimental vulnerability analysis

“Modal Hints” for ManaGPT: Better AI Text Generation Through Prompts Employing the Language of Possibility, Probability, and Necessity

ManaGPT: 4,080 NLP Prompts and Generated Texts

ManaGPT-1020: An LLM Generating Insights into “Posthumanized” Organizations

8-Bit Mystique: An Ingardenian Aesthetic Analysis of the Appeal of Retro Computer Games

A Better Way of Forecasting Employees’ Performance: Evaluating the use of composite ceiling-floor models for predicting the likely range of workers’ future job performance

Optimize Your Performance Intervals! Use ANNs with custom loss functions to predict probable ceilings and floors for workers’ future job performance

One Number Is Rarely Enough: Why prediction intervals are critical (and challenging!) for HR predictive analytics

Comport_AI™

Factory Workers’ Daily Performance & Attrition

Synaptans WorkforceSim™

Revisiting Ingarden’s Theoretical Biological Account of the Literary Work of Art: Is the Computer Game an ‘Organism’?

Ingarden’s Systems-Theoretical Philosophical Anthropology as a Tool for Analyzing the Technological Posthumanization of Persons and Societies

Utopian Confederation: Conspectus

The Self-Revelation and Cognition of VR-Facilitated Virtual Worlds: Towards a Phenomenology of Virtual Habitation

Phenomenology of the Gameworld: A Philosophical Toolbox for Video Game Developers

Beyond Buildings: A Systems-Theoretical Phenomenological Aesthetics of ‘Impossible’ Architectural Structures for Computer Games

Twarde światło i architektura kosmiczna: Analiza fenomenologiczno-estetyczna filmu Tron jako wzorca paradygmatu ‘Siatki Cyberprzestrzennej’

Publications

Presentations

Publications & Presentations by Year