EN

AETERN AI · RESEARCH

Research

Aetern AI’s research focuses on long-term memory, user understanding, identity continuity and memory governance in personal AI. This page presents our publications, core research questions and related literature, spanning representative work from classical philosophy to contemporary research on AI memory. Public-domain texts are available in full on this site; commentary reflects Aetern AI’s interpretation, not the authors’.

PUBLICATIONS

  1. Memory, Outside the ModelAetern AI Research · November 2026Portability and User Control in Personal AIPosition paper
  2. How History Becomes a SelfAetern AI Research · November 2026Personality Formation and Cross-Model Continuity in Personal AIConceptual working paper

QUESTIONS & LITERATURE

R.01 · Long-term memory

How should long-term memory be stored, organized and retrieved?

How interaction records spanning months or years are organized, retrieved and updated.

  1. Theaetetus: The Wax Tablet and the Aviaryc. 369 BCPlato · Theaetetus, translated by Benjamin JowettFull text
  2. Confessions, Book X: The Fields and Spacious Palaces of Memoryc. 400Augustine of Hippo · Confessions, Book X, translated by E. B. PuseyFull text
  3. Why there are complementary learning systems in the hippocampus and neocortex: Insights from the successes and failures of connectionist models of learning and memory1995James L. McClelland et al. · Psychological Review 102(3), 419–457
  4. Memory Networks2015Jason Weston et al. · ICLR 2015
  5. Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks2020Patrick Lewis et al. · NeurIPS 2020
  6. Locating and Editing Factual Associations in GPT2022Kevin Meng et al. · NeurIPS 2022
  7. MemGPT: Towards LLMs as Operating Systems2023Charles Packer et al. · arXiv preprint

R.02 · Understanding users

How does an understanding of the user form from source expressions, and how is it revised?

How understanding is derived from what users actually say, and revised as new information emerges.

  1. The construction of autobiographical memories in the self-memory system2000Martin A. Conway, Christopher W. Pleydell-Pearce · Psychological Review 107(2), 261–288
  2. Generative Agents: Interactive Simulacra of Human Behavior2023Joon Sung Park et al. · UIST 2023
  3. Reflexion: Language Agents with Verbal Reinforcement Learning2023Noah Shinn et al. · NeurIPS 2023
  4. Cognitive Architectures for Language Agents2024Theodore R. Sumers et al. · Transactions on Machine Learning Research (TMLR) 2024
  5. LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memory2025Di Wu et al. · ICLR 2025

R.03 · Identity continuity

What constitutes an AI’s identity across model changes, and how does it persist?

Which attributes should persist as models, applications and devices change, and how to measure them.

  1. Of Identity and Diversity1694John Locke · An Essay Concerning Human Understanding, Book II, Chapter XXVII (added in the second edition)Full text
  2. Of Personal Identity1739David Hume · A Treatise of Human Nature, Book I, Part IV, Section VIFull text
  3. Lessons From an App Update at Replika AI: Identity Discontinuity in Human-AI Relationships2024Julian De Freitas et al. · Harvard Business School Working Paper No. 25-018
  4. “Please, don’t kill the only model that still feels human”: Understanding the #Keep4o Backlash2026Huiqian Lai · CHI 2026

R.04 · Memory governance

What mechanisms should govern forgetting, correction and deletion?

Traceable sources, correctable understanding, verifiable deletion and data portability.

  1. The seven sins of memory: Insights from psychology and cognitive neuroscience1999Daniel L. Schacter · American Psychologist 54(3), 182–203
  2. Machine Unlearning2021Lucas Bourtoule et al. · IEEE Symposium on Security and Privacy 2021
  3. MemoryBank: Enhancing Large Language Models with Long-Term Memory2024Wanjun Zhong et al. · AAAI 2024

METHODOLOGY

  1. Specification as tests

    Key design constraints formalized as automatically verifiable tests.

  2. Fault injection

    Defects deliberately injected before release to verify that tests detect them.

  3. Public benchmarks

    System capabilities evaluated on public long-term memory benchmarks.

  4. Continuity validation

    Consistency across model changes measured against predefined criteria.