hallucinations
Errors where an AI language model generates plausible-sounding but false or fabricated information; Scholl notes engineers must edit and fix these in LLM-generated documents.
As defined in Blake Scholl
Other definitions
Phenomenon where AI systems generate false or fabricated information with confidence, including making up facts, citations, or entire narratives that never occurred or existed. It represents a fundamental safety and reliability problem in current AI systems.
Why Bezos Silenced The Post | EP 107
Where it comes up · 13
- AI Snake Oil: What Artificial Intelligence Can Do, What It Can’t, and How to Tell the Difference5 claims
- This is not the AI we were promised | The Royal Society2 claims
- 317 - What Do We Know About Our Minds2 claims
- Jeffrey Sherman and Jason Draho: How Should I Be Positioned? | UBS On-Air2 claims
- Nobel Prize-Winning Economist Sees Era Of "Permacrisis" Ahead | Michael Spence (w/ Adam Taggart)2 claims
- Robot Plumbers, Robot Armies, and Our Imminent A.I. Future | Interesting Times with Ross Douthat2 claims
- Why Bezos Silenced The Post | EP 1071 claim
- Tyler Cowen on how AI will reorder economies, schools, and spirituality1 claim
and 5 more
Most important claims · 10 of 22Search all claims for “hallucinations” →
- Hallucinations are a fundamental problem with generative AI that will not reach zero in a short period of time; generative AI is stochastic technology that remixes statistical patterns from training data at generation time, so even if training data were perfectly cleaned, hallucinations would persist.AI Snake Oil: What Artificial Intelligence Can Do, What It Can’t, and How to Tell the Difference
- AI hallucinations (where systems make things up) are an industry standard problem, not unique to OpenAI, and companies typically include small disclaimers that their systems can make mistakes rather than solving the underlying problem.Why Bezos Silenced The Post | EP 107
- Generative AI companies have engaged in irresponsible release practices with many harmful consequences, including AI-generated books on Amazon, AI-generated foraging guides full of hallucinations that cause real harm, AI nude-generation apps that have affected hundreds of thousands of people (primarily women), and AI chatbot apps encouraging suicidal ideation.AI Snake Oil: What Artificial Intelligence Can Do, What It Can’t, and How to Tell the Difference
- Large language models have hallucinations (generate false information) because they learn from human-generated data and humans are not entirely accurate; this is acknowledged as a limitation but also reflects the human condition—people transmit and propagate inaccuracies.Ray Kurzweil: Does AI Make Immortality Possible?
- Generative AI has enabled harmful practices at scale through irresponsible release practices, including AI-generated books with hallucinations in critical domains like foraging guides that have caused deaths, nude image generation apps affecting hundreds of thousands primarily of women, and AI-encouraged suicidal behaviors in companion chatbots.AI Snake Oil: What Artificial Intelligence Can Do, What It Can’t, and How to Tell the Difference
- LLMs are powerful for filling out regulatory paperwork with important second-order effects: a certification document like a 100-page lightning-strike test plan that historically took an engineer two months can be produced in minutes via a prompt and a RAG of federal guidance, then edited by a creative engineer—shifting organizations from large armies of crank-turning talent to small teams of creative, change-tolerant talent by reducing the cost of change.Blake Scholl
- Hallucinations are deeply embedded in large language models because they are designed to produce the most plausible, most likely word given training data, not to distinguish truth from falsity, so the model produces high-probability words regardless of accuracy.This is not the AI we were promised | The Royal Society
- ChatGPT produces hallucinations—confident false statements—like claiming Mitchell wrote a book that didn't exist; these happen because the systems are trained to predict plausible text, not to tell the truth, and have no grounding in actual facts.The Future of Artificial Intelligence
- Large language models like ChatGPT have 'hallucinations' (the term used by technologists) where they generate false information as if it were real, as demonstrated by a lawyer who used ChatGPT to prepare a court brief without verification and the AI fabricated legal precedents.Nobel Prize-Winning Economist Sees Era Of "Permacrisis" Ahead | Michael Spence (w/ Adam Taggart)
- Hallucinations in generative AI are a fundamental feature not a bug, arising from the stochastic nature of these models that remix statistical patterns rather than retrieve facts, and will not reach zero rates in the near term despite significant improvements.AI Snake Oil: What Artificial Intelligence Can Do, What It Can’t, and How to Tell the Difference