The Hidden Geometry of AI Language

Action to Language via the Mirror Neuron System, 2006

A large part of the mystery of the origin of language is the difficulty we experience in trying to imagine what the intermediate stages along the way to language could have been. An elegant, detailed, formal account of how discourse interpretation works in terms of a mode of inference called abduction, or inference to the best explanation, enables us to spell out with some precision a quite plausible sequence of such stages. In this chapter I outline plausible sequences for two of the key features of language-Gricean nonnatural meaning and syntax. I then speculate on the time in the evolution of modern humans each of these steps may have occurred.

The Mirror Effect in AI Systems: How Large Language Models Reflect Rather Than Generate Knowledge, 2025

This paper analyzes further the "mirror effect" in Large Language Models (LLMs)-the tendency of AI systems to reflect user expectations and linguistic patterns rather than generate autonomous content. The analysis incorporates historical perspectives on technological mystification, from ancient automata to contemporary AI, and examines transparency deficits in current AI development that obscure these mirroring processes. The goal is to contribute to critical AI literacy by reframing AI capabilities as reflective rather than generative, with significant implications for scientific integrity and epistemological practice.

With the increasing ubiquity of natural language processing (NLP) algorithms, interacting with “conversational artificial agents” such as speaking robots, chatbots, and personal assistants will be an everyday occurrence for most people. In a rather innocuous sense, we can perform a variety of speech acts with them, from asking a question to telling a joke, as they respond to our input just as any other agent would. However, in a stricter, philosophical sense, the question of what we are doing when we interact with these agents is less trivial, as the conversational instances are growing in complexity, interactivity, and anthropomorphic believability. For example, open domain chatbots will soon be capable of holding conversations on a virtually unlimited range of topics. Many philosophical questions are brought up in this development that this special issue aims to address. Are we engaging in a “discourse” when we “argue” with a chatbot? Do they perform speech acts akin to human agents, or do we require different explanations to understand this kind of interactivity? In what way do artificial agents “understand” language, partake in discourse and create text? Are our conversational assumptions andprojections transferable, and should we model artificial speakers along those conventions? Will their moral utterances ever have validity? This special issue of Minds and Machines invites papers discussing a range of topics on human-machine communication and the philosophical presumptions and consequences for developing, distributing, and interacting with speaking robots. We invite the submission of papers focusing on but not restricted to: - What are philosophically sound distinctions between speaking robots, unembodied chatbots, and other forms of artificial speakers? - What constitutes discourse participants, and can artificial speakers ever meet those requirements? - Can artificial speakers perform speech acts, and if yes, can they perform all speech acts humans can perform? Or do robots perform unique speech acts? - What kind of artificial agent can be capable of what kind of language or discourse performance: chatbots, robots, virtual agents,…? - What is the role of anthropomorphism in modelling chatbots as possible discourse participants? - What is the role of technomorphism in modelling human interlocutors as technical discourse participants? - What are the normative consequences of moral statements made by artificial discourse participants? - How will communicative habits between humans change by the presence of artificial speakers? - How can semantic theories explain the meaning-creation of artificial speakers? - Are normative conventions in human-human communication (politeness, compliments) relevant and transformable/transferrable to human-machine communication? - Are there – analogous to human-human communication – any communicative presuppositions in human-machine communication? To submit a paper for this special issue, authors should go to the journal’s Editorial Manager https://www.editorialmanager.com/mind/default.aspx Deadline to submit full paper: October 1st, 2020 First round of reviews: October 2nd – December 1st, 2020 Deadline to resubmit paper: December 15th, 2020 Second round of reviews: December 15th – December 31st, 2020 Deadline for final paper: December 31st, 2020 Publication of special edition: March 2021

Philosophical Investigations, 2025

Abstract: I shall first consider two puzzles that illustrate the contrast between everyday experience or ordinary language, on the one hand, and scientific description on the other. What is common to them is simply that the ordinary description and the scientific description seem to conflict, and the philosopher is called upon to resolve the apparent contradiction. I contend—with some caveats—there is no such conflict, nothing to adjust. That is one philosophical point (which has been made before). The other is to articulate the lesson for a third puzzle, for the concept of intelligence, particularly with respect to AI or Artificial Intelligence (especially as purportedly instantiated by LLMs, ‘Large Language Models’). Keywords: Artificial intelligence, Ordinary Language, Science, Ebersole, Quine

2024 IJRAR March 2024, Volume 11, Issue 1 , 2024

This paper explores the symbiotic relationship between linguistics and artificial intelligence (AI) by investigating their intersection, methodologies, key findings, and broader implications. It addresses how AI techniques can advance linguistic understanding and language-related applications. Employing a comprehensive literature review, it discusses relevant linguistic concepts, AI techniques in linguistic research, and AI-driven linguistic advancements. Ethical considerations regarding AI in linguistics are also covered. The study finds that AI models effectively integrate linguistic knowledge, enhancing applications like machine translation and sentiment analysis. However, challenges like data bias and model complexity are acknowledged. This research emphasizes responsible AI use and highlights the promising future of linguistics and AI collaboration.

Arxiv preprint arXiv:1111.6843, 2011

Barring swarm robotics, a substantial share of current machinehuman and machine-machine learning and interaction mechanisms are being developed and fed by results of agent-based computer simulations, gametheoretic models, or robotic experiments based on a dyadic communication pattern. Yet, in real life, humans no less frequently communicate in groups, and gain knowledge and take decisions basing on information cumulatively gleaned from more than one single source. These properties should be taken into consideration in the design of autonomous artificial cognitive systems construed to interact with//learn from more than one contact or 'neighbor'. To this end, significant practical import can be gleaned from research applying strict science methodology to phenomena humanistic and social, e.g. to discovery of realistic creativity potential spans, or the 'exposure thresholds' after which new information could be accepted by a c ognitive agent. Such rigorous data-driven research offers the chance of not only approximating to descriptive adequacy, but also moving beyond explanatory adequacy to approaching principled explanation. Whether in order to mimic them, or to 'enhance' them, parameters gleaned from complexity science approaches to humans' social and humanistic behavior should subsequently be incorporated as points of reference in the field of robotics and human-machine interaction.

Aamas, 1997

This paper introduces Linguistic Style Improvisation, a theory and set of algorithms for improvisation of spoken utterances by artificial agents, with applications to interactive story and dialogue systems. We argue that linguistic style is a key aspect of character, and show how speech act representations common in AI can provide abstract representations from which computer characters can improvise. We show that the mechanisms proposed introduce the possibility of socially oriented agents, meet the requirements that lifelike characters be believable, and satisfy particular criteria for improvisation proposed by Hayes-Roth. *

2011

This paper investigates the relationship between embodied interaction and symbolic communication. We report about an experiment in which simulated autonomous robotic agents, whose control systems were evolved through an artificial evolutionary process, use abstract communication signals to coordinate their behavior in a context independent way. This use of signals includes some fundamental aspects of sentences in natural languages which are discussed by using the concept of joint attention in relation to the grammatical structure of sentences.

2025

The spontaneous emergence of communication protocols among artificial intelligence agents represents one of the most fascinating phenomena in modern computational systems. This research presents groundbreaking findings on how autonomous AI agents develop sophisticated communication systems without pre-programmed linguistic rules or structures when placed in collaborative environments. Our work reveals that these emergent protocols serve immediate task-specific needs and exhibit characteristics traditionally associated with natural languages, such as compositionality and symbolic abstraction. Through extensive experimental studies conducted over three years, we have identified four distinct phases in evolving communication protocols. The initial phase, ploration phase," involves agents generating and testing random signals within their environment. This is followed by the "signal consolidation phase," where successful communication patterns begin to stabilize through reinforcement. The third "protocol optimization phase" sees the emergence of efficient, minimalist communication structures. Finally, in the "protocol maturation phase," agents develop sophisticated features such as error correction and context-dependent messaging. Our research introduces a novel framework for analyzing these emergent protocols, which we call the Multi-Agent Protocol Evolution Framework (MAPEF). This framework combines elements from information theory, game theory, and linguistic analysis to provide quantitative metrics for evaluating protocol development. Using MAPEF, we demonstrate that the c​o​m​p​

The Pickering and Garrod model (Pickering & Garrod, in press) represents a significant advance within the language-as-action paradigm in providing a mechanistic non-inferential account of dialogue. However, we suggest that, in maintaining several aspects of the language-as-product tradition, it does not go far enough in addressing the dynamic nature of the mechanisms involved. We argue for a radical extension of the language-as-action account, showing how compound-utterance phenomena necessitate a grammar-internal characterization which can only be met with a shift of perspective into one in which linguistic knowledge is seen as procedural.