Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Human-AI Coevolution

Dino Pedreschi, Luca Pappalardo, Emanuele Ferragina, Ricardo Baeza-Yates, Albert-Laszlo Barabasi, Frank Dignum, Virginia Dignum, Tina Eliassi-Rad, Fosca Giannotti, Janos Kertesz, Alistair Knott, Yannis Ioannidis, Paul Lukowicz, Andrea Passarella, Alex Sandy Pentland, John Shawe-Taylor, Alessandro Vespignani

TL;DR

The paper tackles the problem of understanding human-AI feedback loops generated by ubiquitous recommender systems. It introduces Coevolution AI as a field at the crossroads of AI and complexity science, proposing theoretical, empirical, and mathematical investigations of the human-AI loop and outlining methodological gaps. By surveying empirical, simulation, and modelling approaches and detailing outcomes across social media, online retail, urban mapping, and generative AI, it highlights societal impacts such as polarization, inequality, and concentration. The work argues for society-centered AI, data-driven longitudinal studies, and governance mechanisms to steer coevolution toward positive social welfare, offering a roadmap for researchers and policymakers.

Abstract

Human-AI coevolution, defined as a process in which humans and AI algorithms continuously influence each other, increasingly characterises our society, but is understudied in artificial intelligence and complexity science literature. Recommender systems and assistants play a prominent role in human-AI coevolution, as they permeate many facets of daily life and influence human choices on online platforms. The interaction between users and AI results in a potentially endless feedback loop, wherein users' choices generate data to train AI models, which, in turn, shape subsequent user preferences. This human-AI feedback loop has peculiar characteristics compared to traditional human-machine interaction and gives rise to complex and often ``unintended'' social outcomes. This paper introduces Coevolution AI as the cornerstone for a new field of study at the intersection between AI and complexity science focused on the theoretical, empirical, and mathematical investigation of the human-AI feedback loop. In doing so, we: (i) outline the pros and cons of existing methodologies and highlight shortcomings and potential ways for capturing feedback loop mechanisms; (ii) propose a reflection at the intersection between complexity science, AI and society; (iii) provide real-world examples for different human-AI ecosystems; and (iv) illustrate challenges to the creation of such a field of study, conceptualising them at increasing levels of abstraction, i.e., technical, epistemological, legal and socio-political.

Human-AI Coevolution

TL;DR

The paper tackles the problem of understanding human-AI feedback loops generated by ubiquitous recommender systems. It introduces Coevolution AI as a field at the crossroads of AI and complexity science, proposing theoretical, empirical, and mathematical investigations of the human-AI loop and outlining methodological gaps. By surveying empirical, simulation, and modelling approaches and detailing outcomes across social media, online retail, urban mapping, and generative AI, it highlights societal impacts such as polarization, inequality, and concentration. The work argues for society-centered AI, data-driven longitudinal studies, and governance mechanisms to steer coevolution toward positive social welfare, offering a roadmap for researchers and policymakers.

Abstract

Human-AI coevolution, defined as a process in which humans and AI algorithms continuously influence each other, increasingly characterises our society, but is understudied in artificial intelligence and complexity science literature. Recommender systems and assistants play a prominent role in human-AI coevolution, as they permeate many facets of daily life and influence human choices on online platforms. The interaction between users and AI results in a potentially endless feedback loop, wherein users' choices generate data to train AI models, which, in turn, shape subsequent user preferences. This human-AI feedback loop has peculiar characteristics compared to traditional human-machine interaction and gives rise to complex and often ``unintended'' social outcomes. This paper introduces Coevolution AI as the cornerstone for a new field of study at the intersection between AI and complexity science focused on the theoretical, empirical, and mathematical investigation of the human-AI feedback loop. In doing so, we: (i) outline the pros and cons of existing methodologies and highlight shortcomings and potential ways for capturing feedback loop mechanisms; (ii) propose a reflection at the intersection between complexity science, AI and society; (iii) provide real-world examples for different human-AI ecosystems; and (iv) illustrate challenges to the creation of such a field of study, conceptualising them at increasing levels of abstraction, i.e., technical, epistemological, legal and socio-political.
Paper Structure (5 sections, 3 figures)

This paper contains 5 sections, 3 figures.

Table of Contents

  1. Introduction
  2. Methods for human-AI coevolution
  3. Outcomes of human-AI coevolution
  4. Social impact of human-AI coevolution
  5. Open Challenges of Coevolution AI

Figures (3)

  • Figure 1: Users' choices on online platforms generate data used to train recommenders. These recommenders then offer suggestions to users, influencing their choices, which in turn generate more data for re-training recommenders. This iterative process creates a potentially endless feedback loop.
  • Figure 2: We enrich current philosophical perspectives about AI, i.e., technology-centred AI, human-centred AI, and collective intelligence, with a new dimension that can be defined as society-centred AI. Society-centred AI brings three additional elements to the debate.
  • Figure 3: Coevolution AI presents important challenges for the future, that can be conceptualised at increasing levels of abstraction: technical, epistemological, legal and socio-political.