ChatGPT is bullshit
2024/06/01 by Michael Townsen Hicks, James Humphries, Joe Slater · 28 voices · 382 citations
Arts and Humanities · Psychology · Social Sciences · #Cognitive science #Computer science #Epistemology #Epistemology, Ethics, and Metaphysics #Ethics and Social Impacts of AI #Misinformation and Its Impacts #Philosophy #Psychology
paper · pdf · doi:10.1007/s10676-024-09775-5
published in Ethics and Information Technology 26(2) (Springer Science+Business Media)
openalex publication_date 2024/06/01 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/06
Abstract
Abstract Recently, there has been considerable interest in large language models: machine learning systems which produce human-like text and dialogue. Applications of these systems have been plagued by persistent inaccuracies in their output; these are often called “AI hallucinations”. We argue that these falsehoods, and the overall activity of large language models, is better understood as bullshit in the sense explored by Frankfurt (On Bullshit, Princeton, 2005): the models are in an important way indifferent to the truth of their outputs. We distinguish two ways in which the models can be said to be bullshitters, and argue that they clearly meet at least one of these definitions. We further argue that describing AI misrepresentations as bullshit is both a more useful and more accurate way of predicting and discussing the behaviour of these systems.
Cited by
- No Certificate, No Categorical Speech Act: A Brouwerian Assertibility Constraint for Public Reason
- Inductive Risk of AI Hype
- "AI Psychosis" in Context: How Conversation History Shapes LLM Responses to Delusional Beliefs
- Illusions of Understanding from Outsourcing Thinking to LLMs
- Why mental metaphors do not help us understand chatbot mistakes
- Hallucinating with AI: Distributed Delusions and “AI Psychosis”
- Losing Our Voice? Generative AI and the Degradation of Human Expression
- What Does 'Human-Centred AI' Mean?
- Exploring instructional designers' utilization and perspectives on generative AI tools: A mixed methods study
- The différance engine: large language models and poststructuralism
- Efficiency vs. understanding: a critical examination of ChatGPT’s performance in context-sensitive annotation tasks
- Looking Beyond the Hype: Understanding the Effects of AI on Learning
- The New Positivism: A Hegelian Critique
- Big Tech as World Makers: A Research Agenda for International Political Sociology
- Evaluating Metrics for Safety with LLM-as-Judges
- What Kind of Reasoning (if any) is an LLM actually doing? On the Stochastic Nature and Abductive Appearance of Large Language Models
- Knowing Your Uncertainty -- On the application of LLM in social sciences
- Generative Artificial Intelligence in Qualitative Research Methods: Between Hype and Risks?
- Approximating the Mathematical Structure of Psychodynamics
- ‘Appears to be about’: an evaluation of AI-generated metadata quality for community archives
- (A)I Cannot See Them: A Situated Reflection on the Simulation of Historical Figures
- Considering the ethics of large machine learning models in the chemical sciences
- Which educational approaches predict students’ generative AI confidence and responsibility?
- SATISFY EASY LADDERS: accessible machine learning to facilitate analysis of larger sensitive text data samples in criminology
- Large models of what? Mistaking engineering achievements for human linguistic agency
- One Size Fits Some: How Journalistic Roles Shape the Adoption of Generative AI
- Exploring students’ (mis)conceptions about ChatGPT-generated text: A qualitative study
- Bullshit universities: the future of automated education
- How to read a paper involving artificial intelligence (AI)
- Pixels and Predictions: Potential of GPT-4V in Meteorological Imagery Analysis and Forecast Communication
- Stop Saying "AI"
- Control search rankings, control the world: what is a good search engine?
- A Two-Step, Multidimensional Account of Deception in Language Models
- What is a fact in the age of generative AI? Fact-checking as an epistemological lens
- Large Language Models can extract morphological data from taxonomic descriptions, but their stochastic nature makes automation challenging: a test on Australian Asteraceae
- LLMs, Truth, and Democracy: An Overview of Risks
- Using conversational AI to reduce science skepticism
- AI hallucinations can’t be stopped — but these techniques can limit their damage
- KI bei Studierenden: Empirische Erhebung zu Nutzungspraxis, Erwartungen und Haltungen
- Controlling AI Agent Participation in Group Conversations: A Human-Centered Approach
- Generative AI in Self-Directed Learning: a thematic scoping review
- Authenticity controversies: Witness media and generative AI imagery
- Cut the bullshit: why GenAI systems are neither collaborators nor tutors
- Newer, Larger, Better? A Critique of the Unreflective LLM Adoption in Communication Research
- The Ouroboros effect and heterodox domains
- ChatGPT is Bullshit (Partly) Because People are Bullshitters
- Stench of Errors or the Shine of Potential: The Challenge of (Ir)Responsible Use of ChatGPT in Speech‐Language Pathology
- Crisis-line workers’ perspectives on AI in suicide prevention: a qualitative exploration of risk and opportunity
- Generative artificial intelligence (AI) and its performance at indexing tasks
- Psychiatric Risk Associated With Large Language Model Chatbots: An Emerging Concern
- Transforming Agency. On the mode of existence of Large Language Models
- Utopian and dystopian visions: Steering a course for the responsible use of artificial intelligence (AI) in language testing and assessment
- An empirical study of manual abstraction between class diagrams and code of open-source systems
- Learning to Choose or Choosing to Learn: Best-of-N vs. Supervised Fine-Tuning for Bit String Generation
- The limits of large language models and the necessity of human cognition in K-12 education
- Architecting Trust in Artificial Epistemic Agents
- The Limits of Learning from Pictures and Text: Vision-Language Models and Embodied Scene Understanding
- AI systems are not information systems. Should we care?
- Asking an AI for salary negotiation advice is a matter of concern: Controlled experimental perturbation of ChatGPT for protected and non-protected group discrimination on a contextual task with no clear ground truth answers. [europepmc]
- Development of Evidence-Based Guidelines for the Integration of Generative AI in University Education Through a Multidisciplinary, Consensus-Based Approach. [europepmc]
- Intuitive judgements towards artificial intelligence verdicts of moral transgressions. [europepmc]
- Crisis-line workers' perspectives on AI in suicide prevention: a qualitative exploration of risk and opportunity. [europepmc]
- Large Language Models can extract morphological data from taxonomic descriptions, but their stochastic nature makes automation challenging: a test on Australian Asteraceae. [europepmc]
- In Silico Frontiers Shaping the Next Generation of Transformation Product Prediction and Toxicological Assessment. [europepmc]
- Disembodied creativity in generative AI: prima facie challenges and limitations of prompting in creative practice. [europepmc]
- Large language models as cognitive shortcuts: a systems-theoretic reframing beyond bullshit. [europepmc]
- AI-Assisted Rapid Quality Analysis in Implementation Science: Methodological Study. [europepmc]
- What Does 'Human-Centred AI' Mean? [europepmc]
- Development and Initial Testing of an Artificial Intelligence-Based Virtual Reality Companion for People Living with Dementia in Long-Term Care. [europepmc]
- The FHJ debate: Will artificial intelligence replace clinical decision making within our lifetimes? [europepmc]
- Opportunities and challenges of using generative AI to personalize educational assessment. [europepmc]
- Digital Doppelgängers and Lifespan Extension: What Matters? [europepmc]
- Why we need to be careful with LLMs in medicine. [europepmc]
- LLMs, Truth, and Democracy: An Overview of Risks. [europepmc]
Discussions
- How to title a paper. doi.org/10.1007/s106... [bsky, 788 points, 19 comments]
- "ChatGPT is bullshit" doi.org/10.1007/s106... [bsky, 18 points, 1 comments]
- If your tastes run to (slightly) less profanity and (much) more academic peer review, then we got you covered: MH Hicks, J Humphries, and J Slater, "ChatGPT is Bullshit", _Ethics and Info Tech_ 26:38 [bsky, 15 points, 0 comments]
- Reposting with alt text. The DOI referred to in the alt text (and visible at the top of the image) is doi.org/10.1007/s106... [bsky, 12 points, 1 comments]
- 🗒️ Hicks et al. (2024) “ChatGPT Is Bullshit.” Ethics and Information Technology 26 (2): 1–10. doi.org/10.1007/s106.... [bsky, 11 points, 3 comments]
- I love this! I avoid attributing "hallucinations" to GAI since that implies a temporary malfunction rather than a fundamental feature. Instead, I refer to it as a machine designed to produce bullshit, [bsky, 7 points, 1 comments]
- ChatGPT is bullshit. doi.org/10.1007/s106... [bsky, 6 points, 0 comments]
- I liked this take: Hicks, M. T., Humphries, J., & Slater, J. (2024). ChatGPT is bullshit. Ethics and Information Technology, 26(2), 2. doi.org/10.1007/s106... [bsky, 4 points, 0 comments]
- ... my understanding is that that's what GenAI is literally designed to do ("ChatGPT is Bullshit"). ("Bullshit" is a technical term in philosophy - plato.stanford.edu/archIves/fal... - for speech that [bsky, 4 points, 1 comments]
- to use a technical term... doi.org/10.1007/s106... [bsky, 3 points, 0 comments]
- 100% agree on all the things...synthetic text extrusions from LLMs are also properly known as Bullshit. Hicks, M. T., Humphries, J., & Slater, J. (2024). ChatGPT is bullshit. Ethics and Information Te [bsky, 3 points, 1 comments]
- Hicks, M.T., Humphries, J. & Slater, J. ChatGPT is bullshit. Ethics Inf Technol 26, 38 (2024). doi.org/10.1007/s106... [bsky, 3 points, 0 comments]
- Cf doi.org/10.1007/s106... [bsky, 2 points, 1 comments]
- Very similar argument to this fantastic paper doi.org/10.1007/s106... [bsky, 2 points, 0 comments]
- Reminder: "bullshit" is a formally defined philosophical categorization. ChatGPT is bullshit (Townsen Hicks, 2024) doi.org/10.1007/s106... [bsky, 2 points, 1 comments]
- doi.org/10.1007/s106... [bsky, 1 points, 0 comments]
- Tapping the sign! doi.org/10.1007/s106... [bsky, 1 points, 0 comments]
- Don't you know that AI is bullshit?* *that's actually the academic term. See Hicks et al. (2024): doi.org/10.1007/s106... [bsky, 1 points, 1 comments]
- In this context, highly mandatory reading: doi.org/10.1007/s106... [bsky, 1 points, 1 comments]
- ChatGPT is bullshit - Ethics and Information Technology https:// link.springer.com/article/10.1 007/s10676-024-09775-5 What a great read. # OpenAccess [mastodon, 0 points, 1 comments]
- doi.org/10.1007/s106... [bsky, 0 points, 0 comments]
- Not exactly news but nice to see it in an academic paper. [bsky, 0 points, 0 comments]
- Here's the paper that makes the argument: doi.org/10.1007/s106... (Name of journal D.A.C...) [bsky, 0 points, 1 comments]
- & this? Hicks, Michael Townsen, James Humphries, and Joe Slater. ‘ChatGPT Is Bullshit’. Ethics and Information Technology 26, no. 2 (2024): 38. doi.org/10.1007/s106.... [bsky, 0 points, 0 comments]
- ChatGPT is bullshit. Like, scientifically speaking. 🧪 doi.org/10.1007/s106... [bsky, 0 points, 0 comments]
- There are good technical accounts of why the basic architecture of what we call an LLM cannot do this. doi.org/10.1007/s106... doi.org/10.1007/978-... [bsky, 0 points, 0 comments]
- I hope you will give careful consideration to: Hicks, M.T., Humphries, J. & Slater, J. ChatGPT is bullshit. Ethics Inf Technol 26, 38 (2024). doi.org/10.1007/s106... [bsky, 0 points, 0 comments]
- If ChatGPT and other LLMs are "bullshit-machines", as arguably they are, however crass the term, how shall we think of vision transformers and especially their output in scientific image analysis? [bsky, 0 points, 0 comments]
Related