The Surprising Creativity of Digital Evolution: A Collection of Anecdotes from the Evolutionary Computation and Artificial Life Research Communities
Not on the map. Why?
Abstract (arXiv)
Biological evolution provides a creative fount of complex and subtle adaptations, often surprising the scientists who discover them. However, because evolution is an algorithmic process that transcends the substrate in which it occurs, evolution's creativity is not limited to nature. Indeed, many researchers in the field of digital evolution have observed their evolving algorithms and organisms subverting their intentions, exposing unrecognized bugs in their code, producing unexpected adaptations, or exhibiting outcomes uncannily convergent with ones in nature. Such stories routinely reveal creativity by evolution in these digital worlds, but they rarely fit into the standard scientific narrative. Instead they are often treated as mere obstacles to be overcome, rather than results that warrant study in their own right. The stories themselves are traded among researchers through oral tradition, but that mode of information transmission is inefficient and prone to error and outright loss. Moreover, the fact that these stories tend to be shared only among practitioners means that many natural scientists do not realize how interesting and lifelike digital organisms are and how natural their evolution can be. To our knowledge, no collection of such anecdotes has been published before. This paper is the crowd-sourced product of researchers in the fields of artificial life and evolutionary computation who have provided first-hand accounts of such cases. It thus serves as a written, fact-checked collection of scientifically important and even entertaining stories. In doing so we also present here substantial evidence that the existence and importance of evolutionary surprises extends beyond the natural world, and may indeed be a universal property of all complex evolving systems.
Key facts. Cited by 78 papers in the map (data through 14 Oct 2026). First posted on arXiv 9 Mar 2018.
Builds on
No reference list yet.
Built on by
Citing papers per year: 2018: 2; 2019: 0; 2020: 0; 2021: 8; 2022: 14; 2023: 13; 2024: 16; 2025: 12; 2026: 13.
78 papers.
Sort:
By area:
- 2022Scaling Laws for Reward Model Overoptimizationcited by 1,095
- 2018Scalable agent alignment via reward modeling: a research directioncited by 447
- 2018Reward learning from human preferences and demonstrations in Ataricited by 316
- 2021Unsolved Problems in ML Safetycited by 292
- 2022The Alignment Problem from a Deep Learning Perspectivecited by 258
- 2023Reward Model Ensembles Help Mitigate Overoptimizationcited by 231
- 2023AI Deception: A Survey of Examples, Risks, and Potential Solutionscited by 188
- 2023Guiding Pretraining in Reinforcement Learning with Large Language Modelscited by 187
- 2021Alignment of Language Agentscited by 168
- 2022Toward Transparent AI: A Survey on Interpreting the Inner Structures of Deep Neural Networkscited by 129
- 2023Training Socially Aligned Language Models on Simulated Social Interactionscited by 78
- 2022On scientific understanding with artificial intelligencecited by 70
38 more
- 2023OMNI: Open-endedness via Models of human Notions of Interestingnesscited by 70
- 2024Uncovering Deceptive Tendencies in Language Models: A Simulated Company AI Assistantcited by 38
- 2023Emergence of Maps in the Memories of Blind Navigation Agentscited by 33
- 2024Rethinking Reward Model Evaluation: Are We Barking up the Wrong Tree?cited by 31
- 2022Reward Gaming in Conditional Text Generationcited by 30
- 2021Human irrationality: both bad and good for reward inferencecited by 25
- 2022Covariance Matrix Adaptation MAP-Annealingcited by 25
- 2022Mind the gap: Challenges of deep learning approaches to Theory of Mindcited by 21
- 2023DITTO: Offline Imitation Learning with World Modelscited by 21
- 2024A social path to human-like artificial intelligencecited by 20
- 2024Language Models can Subtly Deceive Without Lying: A Case Study on Strategic Phrasing in Legislationcited by 10
- 2021Definitions of intent suitable for algorithmscited by 8
- 2024Text2Robot: Evolutionary Robot Design from Text Descriptionscited by 8
- 2024From Text to Life: On the Reciprocal Relationship between Artificial Life and Large Language Modelscited by 8
- 2025Enhancing Large Language Model Reasoning with Reward Models: An Analytical Surveycited by 8
- 2025Alignment, Agency and Autonomy in Frontier AI: A Systems Engineering Perspectivecited by 7
- 2021Understanding and Avoiding AI Failures: A Practical Guidecited by 6
- 2023Bias Amplification Enhances Minority Group Performancecited by 6
- 2025AI Testing Should Account for Sophisticated Strategic Behaviourcited by 6
- 2023Evolving Populations of Diverse RL Agents with MAP-Elitescited by 5
- 2025Evolvable Conditional Diffusioncited by 5
- 2022Severe Damage Recovery in Evolving Soft Robots through Differentiable Programmingcited by 4
- 2022Towards Explainable Metaheuristic: Mining Surrogate Fitness Models for Importance of Variablescited by 4
- 2025RIVAL: Reinforcement Learning with Iterative and Adversarial Optimization for Machine Translationcited by 4
- 2021Carle's Game: An Open-Ended Challenge in Exploratory Machine Creativitycited by 3
- 2022Evolved Open-Endedness in Cultural Evolution: A New Dimension in Open-Ended Evolution Researchcited by 3
- 2023Machine Lovecited by 3
- 2022EvolvingBehavior: Towards Co-Creative Evolution of Behavior Trees for Game NPCscited by 2
- 2023CERN for AI: A Theoretical Framework for Autonomous Simulation-Based Artificial Intelligence Testing and Alignmentcited by 2
- 2024Quantifying stability of non-power-seeking in artificial agentscited by 2
- 2024Seizing the Means of Production: Exploring the Landscape of Crafting, Adapting and Navigating Generative AI Models in the Visual Artscited by 2
- 2024Evolutionary Computation and Explainable AI: A Roadmap to Understandable Intelligent Systemscited by 2
- 2025What if Eye...? Computationally Recreating Vision Evolutioncited by 2
- 2021Quality Evolvability ES: Evolving Individuals With a Distribution of Well Performing and Diverse Offspringcited by 1
- 2022Qualitative Differences Between Evolutionary Strategies and Reinforcement Learning Methods for Control of Autonomous Agentscited by 1
- 2024On Creativity and Open-Endednesscited by 1
- 2024AI-Guided Codesign Framework for Novel Material and Device Design applied to MTJ-based True Random Number Generatorscited by 1
- 2025Winning at All Cost: A Small Environment for Eliciting Specification Gaming Behaviors in Large Language Modelscited by 1
50 most cited of 78 All 78 (JSON)
- 2026Directing large language models to follow the letter or spirit of the law19 Sep 2026
- 2026Gaming Without an Attacker: Benchmark Fingerprinting in LLM-Driven Search Under Selection Pressure9 Aug 2026
- 2026Fragility of Value under Imperfect Alignment30 Jul 2026
- 2026Win by Silence: Deletion Non-Monotonicity, Autonomous Exploitation, and Typed-State Gating in LLM Plan Evaluation14 Jul 2026
- 2026Sample Efficient Generative Optimization for Molecular Design14 Jul 2026
- 2026Conversable Complexity: Agentic LLM Collectives as Interpretable Substrates1 Jul 2026
- 2026Knowing in Advance When an Evolutionary Outer Loop Will Not Help: A Pre-Registered Cheap-Baseline Screening Rule28 Jun 2026
- 2026Greed Is Learned: Visible Incentives as Reward-Hacking Triggers15 Jun 2026
- 2026Signed Compression Progress on a Sealed Audit is Goodhart-Resistant9 Jun 2026
- 2026The Time is Here for Just-in-Time Systems: Challenges and Opportunities22 May 2026
- 2026Questionnaire Responses Do not Capture the Safety of AI Agents15 Mar 2026
- 2026HCP-DCNet: A Hierarchical Causal Primitive Dynamic Composition Network for Self-Improving Causal Understanding12 Mar 2026
38 more
- 2026Status Hierarchies in Language Models24 Jan 2026
- 2025On the Computability of Artificial General Intelligence4 Dec 2025
- 2025Evolutionary Computation as Natural Generative AI4 Oct 2025
- 2025Enhancing Large Language Model Reasoning with Reward Models: An Analytical Survey2 Oct 2025
- 2025AI Testing Should Account for Sophisticated Strategic Behaviour19 Aug 2025
- 2025Towards Integrated Alignment8 Aug 2025
- 2025Evolvable Conditional Diffusion16 Jun 2025
- 2025RIVAL: Reinforcement Learning with Iterative and Adversarial Optimization for Machine Translation5 Jun 2025
- 2025Serious Games: Human-AI Interaction, Evolution, and Coevolution22 May 2025
- 2025Winning at All Cost: A Small Environment for Eliciting Specification Gaming Behaviors in Large Language Models7 May 2025
- 2025Alignment, Agency and Autonomy in Frontier AI: A Systems Engineering Perspective20 Feb 2025
- 2025Superhuman Game AI Disclosure: Expertise and Context Moderate Effects on Trust and Fairness31 Jan 2025
- 2025What if Eye...? Computationally Recreating Vision Evolution25 Jan 2025
- 2024AI-Guided Codesign Framework for Novel Material and Device Design applied to MTJ-based True Random Number Generators1 Nov 2024
- 2024Unlocking the Potential of Global Human Expertise31 Oct 2024
- 2024Rethinking Reward Model Evaluation: Are We Barking up the Wrong Tree?8 Oct 2024
- 2024Aligning Robot Navigation Behaviors with Human Intentions and Preferences16 Sep 2024
- 2024Text2Robot: Evolutionary Robot Design from Text Descriptions28 Jun 2024
- 2024From Text to Life: On the Reciprocal Relationship between Artificial Life and Large Language Models14 Jun 2024
- 2024Evolutionary Computation and Explainable AI: A Roadmap to Understandable Intelligent Systems12 Jun 2024
- 2024On Creativity and Open-Endedness28 May 2024
- 2024A social path to human-like artificial intelligence22 May 2024
- 2024Language Models can Subtly Deceive Without Lying: A Case Study on Strategic Phrasing in Legislation7 May 2024
- 2024Seizing the Means of Production: Exploring the Landscape of Crafting, Adapting and Navigating Generative AI Models in the Visual Arts26 Apr 2024
- 2024Uncovering Deceptive Tendencies in Language Models: A Simulated Company AI Assistant25 Apr 2024
- 2024Business models for the simulation hypothesis13 Apr 2024
- 2024Evo* 2023 -- Late-Breaking Abstracts Volume20 Mar 2024
- 2024Extinction Risks from AI: Invisible to Science?2 Feb 2024
- 2024Quantifying stability of non-power-seeking in artificial agents7 Jan 2024
- 2023CERN for AI: A Theoretical Framework for Autonomous Simulation-Based Artificial Intelligence Testing and Alignment14 Dec 2023
- 2023Replication of Multi-agent Reinforcement Learning for the "Hide and Seek" Problem9 Oct 2023
- 2023Reward Model Ensembles Help Mitigate Overoptimization4 Oct 2023
- 2023Bias Amplification Enhances Minority Group Performance13 Sep 2023
- 2023AI Deception: A Survey of Examples, Risks, and Potential Solutions28 Aug 2023
- 2023OMNI: Open-endedness via Models of human Notions of Interestingness2 Jun 2023
- 2023Training Socially Aligned Language Models on Simulated Social Interactions26 May 2023
- 2023Efficient Quality Diversity Optimization of 3D Buildings through 2D Pre-optimization28 Mar 2023
- 2023Evolving Populations of Diverse RL Agents with MAP-Elites9 Mar 2023
50 newest of 78 All 78 (JSON)
- 2022Scaling Laws for Reward Model Overoptimizationcited by 1,095
- 2018Scalable agent alignment via reward modeling: a research directioncited by 447
- 2018Reward learning from human preferences and demonstrations in Ataricited by 316
- 2023Reward Model Ensembles Help Mitigate Overoptimizationcited by 231
- 2023Training Socially Aligned Language Models on Simulated Social Interactionscited by 78
- 2025RIVAL: Reinforcement Learning with Iterative and Adversarial Optimization for Machine Translationcited by 4
- 2025Towards Integrated Alignmentcited by 0
- 2026Fragility of Value under Imperfect Alignmentcited by 0
- 2026Fragility of Value under Imperfect Alignment30 Jul 2026
- 2025Towards Integrated Alignment8 Aug 2025
- 2025RIVAL: Reinforcement Learning with Iterative and Adversarial Optimization for Machine Translation5 Jun 2025
- 2023Reward Model Ensembles Help Mitigate Overoptimization4 Oct 2023
- 2023Training Socially Aligned Language Models on Simulated Social Interactions26 May 2023
- 2022Scaling Laws for Reward Model Overoptimization19 Oct 2022
- 2018Scalable agent alignment via reward modeling: a research direction19 Nov 2018
- 2018Reward learning from human preferences and demonstrations in Atari15 Nov 2018
- 2023DITTO: Offline Imitation Learning with World Modelscited by 21
- 2023DITTO: Offline Imitation Learning with World Models6 Feb 2023
- 2021Unsolved Problems in ML Safetycited by 292
- 2022The Alignment Problem from a Deep Learning Perspectivecited by 258
- 2023AI Deception: A Survey of Examples, Risks, and Potential Solutionscited by 188
- 2023Guiding Pretraining in Reinforcement Learning with Large Language Modelscited by 187
- 2021Alignment of Language Agentscited by 168
- 2022Toward Transparent AI: A Survey on Interpreting the Inner Structures of Deep Neural Networkscited by 129
- 2022On scientific understanding with artificial intelligencecited by 70
- 2023OMNI: Open-endedness via Models of human Notions of Interestingnesscited by 70
- 2024Uncovering Deceptive Tendencies in Language Models: A Simulated Company AI Assistantcited by 38
- 2023Emergence of Maps in the Memories of Blind Navigation Agentscited by 33
- 2022Reward Gaming in Conditional Text Generationcited by 30
- 2021Human irrationality: both bad and good for reward inferencecited by 25
12 most cited of 64 All 64 (JSON)
- 2026Directing large language models to follow the letter or spirit of the law19 Sep 2026
- 2026Gaming Without an Attacker: Benchmark Fingerprinting in LLM-Driven Search Under Selection Pressure9 Aug 2026
- 2026Win by Silence: Deletion Non-Monotonicity, Autonomous Exploitation, and Typed-State Gating in LLM Plan Evaluation14 Jul 2026
- 2026Sample Efficient Generative Optimization for Molecular Design14 Jul 2026
- 2026Knowing in Advance When an Evolutionary Outer Loop Will Not Help: A Pre-Registered Cheap-Baseline Screening Rule28 Jun 2026
- 2026Greed Is Learned: Visible Incentives as Reward-Hacking Triggers15 Jun 2026
- 2026Signed Compression Progress on a Sealed Audit is Goodhart-Resistant9 Jun 2026
- 2026The Time is Here for Just-in-Time Systems: Challenges and Opportunities22 May 2026
- 2026HCP-DCNet: A Hierarchical Causal Primitive Dynamic Composition Network for Self-Improving Causal Understanding12 Mar 2026
- 2026Status Hierarchies in Language Models24 Jan 2026
- 2025On the Computability of Artificial General Intelligence4 Dec 2025
- 2025Evolutionary Computation as Natural Generative AI4 Oct 2025
12 newest of 64 All 64 (JSON)