Research programme

We are founding a new field: Computational Personality Science.

A startup that will be acquired in five years builds a product. A startup that will define a category for the next twenty years founds an academic discipline. We are doing the second.

A discipline, not a feature

Computational Personality Science sits at the intersection of four traditions.

Cognitive science

Kahneman’s dual-process theory, Tulving’s episodic/semantic distinction, Vygotsky’s cultural scaffolding. We borrow constructs and contribute longitudinal multimodal data.

Psychometrics & decision science

Schwartz values inventory, Big Five, Russell’s circumplex, Martin’s humor styles. We operationalize them in a single Pydantic schema.

Machine learning & NLP

Foundation models, PEFT/LoRA, federated learning, stylometry, behavioral prompting. We add the structured intermediate that makes them auditable.

Digital humanities & ethics

Ricœur’s narrative identity, Nissenbaum’s contextual integrity, Sofka’s grief tech research. We translate them into architectural primitives.

Open benchmark

The Personality Test™

The Turing Test asks: can a machine be mistaken for a human?That question is no longer interesting. The new question — and the one this category requires — is: can the system be mistaken for a specific human?

The protocol

  1. The creator (a real person) answers 30 unseen questions in writing — the Personality Probe Set (PPS) — held out from training data.
  2. The capsule, built only from the creator’s training-set data, answers the same 30 questions independently.
  3. 3–5 evaluators per creator (typically family members) see paired (real, capsule) answers in random order.
  4. Each pair is rated on three 7-point Likert scales: authenticity, reasoning fidelity, linguistic signature.
  5. Forced-choice distinguishability: which answer was the AI? Random performance = 50%; trivial detection = 100%.
Held-out integrity

The harness in packages/personality/aitc_personality/eval.py refuses to run if any held-out content’s SHA-256 appears in the training corpus index — preventing accidental leakage.

Targets by sprint

SprintCapabilityTarget D
Sprint 1PV + Behavioral Prompting on Claude/GPT-4o≤ 0.75
Sprint 2+ voice (ElevenLabs), + UI-side Authenticity Seal≤ 0.70
Sprint 3+ LoRA on open-source LLM, + biometric inputs≤ 0.65
Sprint 4+ 3D avatar (text-channel evaluation)≤ 0.60
Funded MVPFull system, 50-creator dataset≤ 0.58 + paper

Distinguishability rates around 0.6 are remarkable for cognitive tasks of this kind: even people who know the creator extremely well struggle to tell the capsule apart from the real person.

Public release plan
  • • Protocol v1.0 + 5-creator pilot results — 2026-Q3 — PLOS ONE / CHI 2026
  • • Open-source eval harness — 2026-Q3 — this repository
  • • Public leaderboard — 2026-Q4 — aitimecapsule.com/benchmark
  • • 50-creator funded study — 2027-Q2 — Science Advances target
Manuscript pipeline

Eleven manuscripts in active preparation, lead author Rodion Sorokin.

These are not abstracts. They are full manuscripts (DOCX/PDF, 20–50 pages each) authored over the second half of 2025 and Q1 2026, currently in submission to PLOS ONE, CHI 2026, JMIR, and the ACM Journal on Responsible Computing.

01
Federated Personalization for Scalable Personality Simulation
02
Federated Longitudinal Studies (FLS): A Privacy-by-Design Methodology for Ecologically Valid, Multi-Modal Behavioral Data Collection in the Wild
03
Beyond the Turing Test: A Framework for the Personality Test to Evaluate the Authenticity of a Specific Simulated Identity
04
The Digital Will: A Blockchain-Based Governance Framework for Post-Mortem Sovereignty of a Digital Personality
05
From the Archeology of Memory to the Simulation of Personality: A New Paradigm for Digital Legacy and Narrative Identity
06
Crossing the Uncanny Valley through Behavioral Authenticity: Synthesizing Photorealistic Avatars from Multi-Modal Personality Models
07
A Simulated Dialogue: The Therapeutic Potential and Ethical Considerations of Generative Personality Avatars in Grief Counseling
08
Preserving Intangible Heritage: Using Personality Simulation to Create Dynamic, Interactive Archives of Cultural and Linguistic Knowledge
09
Behavioral Prompting: A Mechanism for Injecting Cognitive Style and Affective Patterns into Large Language Models
10
Foundations of Computational Personality Science: A Manifesto for a New Interdisciplinary Field
11
The Mind-Soul Architecture: A Cognitively Plausible Hybrid for Personalized AI at Population Scale
Academic partnership program

Four research tracks, four target departments at Columbia and beyond.

Decision Science
Columbia Business School

Longitudinal Laboratory for Decision Science — first opportunity to capture lifetime longitudinal data on how decision-making heuristics evolve. Funding: NSF, Sloan.

Bridging Cognitive Science and Clinical Application
Psychology, Psychiatry & Public Health

Empirical validation of therapeutic potential in grief support and the study of cognitive aging. Funding: NIH, private mental-health foundations. IRB-gated throughout.

Responsible AI Governance and Digital Provenance
Computer Science & Journalism School

Living case study for the Columbia Initiative on AI and Data Science for Society (AIDS4S). Joint AI Governance Case Study. Funding: NSF, Knight, DARPA.

The Next Generation of the Archive
Faculties of Arts and Sciences (Humanities)

Partnership with Columbia’s Oral History Archives to create the first digital avatars of historical figures. Funding: Mellon, NEH.

Why the academic flywheel matters strategically

The category will have an academic discipline whether we name it or not. By naming it first, defining its methods first, publishing its benchmark first, and partnering with the most credible institutions first, we make ourselves the de facto reference architecture. This is a moat no startup capital can buy after the fact.