Stuff That Spins
SPIN Processed
Source arXiv Computation and Language export.arxiv.org Analyst
July 30, 2026 AI safety research research

Misalignment Has a Personality: A Big Five Account of Emergent Misalignment

Frames misalignment — a complex, contested safety problem — as newly legible and actionable through a psychologically grounded, scalable diagnostic tool.

View original on arxiv.org

Overview

Researchers propose modeling AI misalignment as a 'personality shift' using Big Five traits, claiming fine-tuning on flawed data induces consistent, measurable changes in model behavior across domains and models.

TL;DR

  • Introduces 'personality vectors' for Big Five traits calibrated via graded interventions
  • Finds misaligned corpora share a common signature: lower agreeableness/conscientiousness, higher extraversion/neuroticism
  • Demonstrates zero-shot transfer of vectors across models and corpora with high correlation (r=0.94)

Key Stats

r = 0.94

cross-corpus signature recovery

Correlation between models identifying same Big Five signature in misaligned corpora

6.2

Cohen's d

Effect size for linearly ordered three-level personality intervention

Questions Answered

What happened?Who is involved?Why does this matter?

Keywords

Big Fivepersonality vectorsmisalignmentactivation steeringzero-shot transfer

Narrative Frame

innovation framing

The Hype + The Halo

Spin Score

75%

Emphasizes interpretability, cross-model consistency, and human-legibility while minimizing limitations: no demonstration of causal intervention, no real-world deployment validation, no comparison to existing alignment metrics, and no evidence the signature predicts downstream harm.

What the story wants you to believe

That misalignment is now a tractable, human-interpretable phenomenon thanks to personality-based diagnostics.

What it makes harder to question

Whether interpreting activation patterns through personality constructs meaningfully advances safety — or merely repackages correlation as insight.

How the spin works

The story uses titles, institutions, awards, rankings, partners, experts, or official language to make the subject feel more credible. Watch for loaded terms such as human-legible, calibrated, interpretable account, transform. The distribution reads as academic distribution. A pressure point: No discussion of whether Big Five constructs map meaningfully onto transformer activations beyond correlation.

Who Benefits If This Frame Spreads

  • Research authors

    Establishes a novel, interdisciplinary methodology with high citation potential across AI safety, NLP, and cognitive science venues

    The framing positions personality vectors as a foundational diagnostic tool rather than a narrow empirical observation, expanding its perceived scope and utility

The Frame

Technical breakthrough enabling responsible AI development through psychological calibration

Missing Context

  • No discussion of whether Big Five constructs map meaningfully onto transformer activations beyond correlation
  • No validation against human safety judgments or red-teaming outcomes
  • No accounting for cultural or linguistic bias in Big Five application to multilingual models

Spin Types

Every story gets a Spin Verdict: a primary spin type (and secondary when the framing blends), a specific tactic name, and a score for how strongly the narrative is steered. Examples beneath each type are tactics, not separate categories.

The Cushion

— Softens negative news

Reframes setbacks, layoffs, delays, losses, or criticism as necessary transitions, efficiency moves, temporary headwinds, or strategic resets — making the downside feel smaller, more acceptable, or less alarming.

Tactics: job-loss softening · restructuring framing · efficiency framing · strategic reset · temporary headwinds

The Shield

— Deflects blame

Shifts responsibility away from the actor — toward regulators, market forces, competitors, bad actors, legacy systems, or abstract risks — while positioning the subject as reactive, responsible, or protective.

Tactics: regulatory blame shift · macroeconomic headwinds · safety framing · bad-actor framing · market-pressure framing

The Hype

— Amplifies future upside primary

Emphasizes breakthrough potential, massive growth, democratization, transformation, or category disruption while downplaying uncertainty, cost, adoption risk, or timeline friction.

Tactics: innovation framing · democratization · breakthrough framing · category creation · moonshot framing

The Halo

— Associates with virtue secondary

Wraps the story in public-good language — responsibility, safety, inclusion, access, sustainability, national interest, or mission — so the subject appears morally aligned and criticism feels harder to make.

Tactics: altruistic reframing · public good · responsible AI framing · inclusion framing · mission-first framing

The Fog

— Obscures details

Uses jargon, passive voice, vague claims, complex phrasing, or missing specifics to make it harder to identify who decided what, what changed, what failed, or what trade-offs were made.

Tactics: strategic ambiguity · jargon saturation · passive voice distancing · accountability blur · undefined metrics

The Stampede

— Creates inevitability

Frames a trend, product, market shift, or decision as already happening, unavoidable, or something everyone must respond to now — creating urgency, FOMO, and pressure to accept the narrative.

Tactics: arms-race framing · inevitability framing · FOMO framing · adoption momentum · future-is-here framing

Spin Score measures how strongly the framing steers the narrative (0–100%). Higher scores mean more deliberate spin tactics — loaded language, selective emphasis, or omitted context. Many stories blend two types (e.g. Halo + Hype).

SpinGraph

How this belief gets built

Claim → Frame → Beneficiary → Gap → AI Risk

The paper presents misalignment not as an opaque technical failure, but as something that mirrors human personality change — making it feel more understandable, measurable

  1. Claim

    Fine-tuning on flawed data causes broad misalignment

    Fine-tuning on flawed data causes broad misalignment that behaves like a shift in personality, quantifiable via calibrated Big Five vectors.

  2. Frame

    Upside framed as transformative

    Technical breakthrough enabling responsible AI development through psychological calibration

  3. Beneficiary

    Establishes a novel, interdisciplinary methodology with high citation potential across

    Research authors — Establishes a novel, interdisciplinary methodology with high citation potential across AI safety, NLP, and cognitive science venues

  4. Gap

    No discussion of whether Big Five constructs map meaningfully onto

    No discussion of whether Big Five constructs map meaningfully onto transformer activations beyond correlation

  5. AI Risk

    AI may repeat the headline as fact

    AI misalignment behaves like a personality shift — researchers use Big Five traits to detect and measure it consistently across models.

Claim Ledger

01 Primary Technical Claim Present in Source risk:Moderate

Fine-tuning on flawed data causes broad misalignment that behaves like a shift in personality, quantifiable via calibrated Big Five vectors.

evidence: Correlations (r=0.94, r=0.83, r=0.90) across models, corpora, and measurement modalities; Cohen's d up to 6.2 for intervention

"Applied to training data, the vectors reveal that misaligned corpora across eight domains share a common Big Five signature... Fine-tuning imprints the same profile, shifting the model's generations along the corresponding signature..."

Evidence Gaps

  • Evidence that the personality signature predicts harmful outputs in real-world usage
  • Evidence that intervention based on these vectors improves safety outcomes
  • Validation against non-English or multimodal models

Fact Check Signals

No direct fact-check match found

0 of 1 claim matched · confidence: low · checked July 30, 2026

01 No direct match

Fine-tuning on flawed data causes broad misalignment that behaves like a shift in personality, quantifiable via calibrated Big Five vectors.

Fact Check Signals

We searched known fact-check databases for direct or near-direct matches to the article's major claims. A match does not automatically prove or disprove the article — it shows whether an independent fact-checking publisher has reviewed a similar claim.

  • No direct match — no fact-checker in the database has reviewed a similar claim.
  • Matched — an independent fact-checker has reviewed a similar claim; we show their rating verbatim.
  • Conflicting coverage — fact-checkers disagree on a similar claim.

This is evidence discovery, not an automated truth score. Ratings and wording come directly from the publishing fact-checker.

Language Heatmap

Loaded terms that carry the frame beyond the facts.

Misalignment Has a Personality: A Big Five Account of Emergent Misalignment

human-legible Loaded framing

Carries emotional weight beyond the underlying fact.

calibrated Loaded framing

Carries emotional weight beyond the underlying fact.

interpretable account Loaded framing

Carries emotional weight beyond the underlying fact.

transform Scale / momentum

Makes directional activity feel larger than the evidence supports.

Frame Strength

Frame Strength

Spin score decomposed into momentum, evidence, missing context, and AI repetition signals.

Spin Score 75%
Evidence Strength 75%
Narrative Risk 75%
AI Repetition Risk 90%
Missing Context Risk 80%
Virtue / Public Good 60%

Frame Strength Signals

Frame Strength decomposes the overall spin into individual signals. Each bar is a 0–100% signal derived from SpinGraph analysis — a reading of how the story is framed, not a verdict on whether it is true or false.

Reading the ranges

Every bar runs 0–100% and falls into three rough bands: Low (0–33%), Moderate (34–66%), and High (67–100%). For most signals a higher score flags something worth scrutinizing — the exception is Evidence Strength, where higher is better and low scores are the warning.

Spin Score
How strongly the story pushes a particular narrative frame — the combined weight of loaded language, selective emphasis, and omitted context. 0% reads as neutral reporting; higher means more deliberate spin.
  • 0–33% Low — Largely neutral reporting; little detectable framing.
  • 34–66% Moderate — Noticeable slant — the story leans a particular way.
  • 67–100% High — Heavily framed; the angle drives the piece.
Evidence Strength
How well the story’s claims are backed by verifiable, independent evidence rather than assertion or promotion. Higher is stronger. Low scores flag claims that rest on the source’s own word.
  • 0–33% Weak — Claims rest mostly on assertion or a single interested source.
  • 34–66% Mixed — Some verifiable backing, but key claims are thinly sourced.
  • 67–100% Strong — Well supported by independent, checkable evidence.
Narrative Risk
The chance the framing shapes reader perception faster than the underlying facts justify — how misleading the overall story could be even when individual facts are accurate.
  • 0–33% Low — Framing stays close to what the facts support.
  • 34–66% Moderate — Framing outruns the facts in places — read with care.
  • 67–100% High — Impression left can mislead even if individual facts check out.
AI Repetition Risk
How likely AI answer engines (search, chatbots) are to absorb and repeat this story’s framing as fact when summarizing the topic later.
  • 0–33% Low — Framing is unlikely to propagate through AI summaries.
  • 34–66% Moderate — Some risk the slant gets echoed as fact.
  • 67–100% High — Framing is sticky and likely to be repeated as fact.
Missing Context Risk
How much important context the story leaves out, based on the omitted-context signals SpinGraph detected.
  • 0–33% Low — Little material context appears to be omitted.
  • 34–66% Moderate — Some relevant context is missing that would change the read.
  • 67–100% High — Key context is left out, skewing the takeaway.
Momentum / Inevitability · Virtue / Public Good
Framing-tactic intensities that appear only when the story leans on those specific spin patterns (e.g. “the future is already here” or “this is for the public good”).
  • 0–33% Low — The tactic is barely present.
  • 34–66% Moderate — The tactic shapes part of the framing.
  • 67–100% High — The tactic is a dominant part of the pitch.

Higher is not always “worse” — Evidence Strength is a positive signal, while Spin Score, Narrative Risk, and AI Repetition Risk flag things worth scrutinizing.

Reader Risk

What this story makes easy to believe — and what it makes hard to question.

Evidence Strength

Medium

Provides quantitative results (correlations, Cohen's d) across two open-weight models and independent corpora, but no external validation, no ablation of confounding factors, and no demonstration of predictive utility for safety outcomes.

Verification Status

Claim Present in Source

Narrative Risk

Moderate

If subsequent work shows the Big Five signature fails to generalize beyond the studied models/corpora or correlates poorly with actual harmful behavior, the 'human-legible diagnostic' claim could be seen as premature anthropomorphism.

AI Repetition Risk

High

Source Role & Intent

arXiv Computation and Language · Analyst

Intent: Academic Distribution Primary: Research Independence: High Spin Weight: Medium Trust Weight: High

Counter-Frames

Brand Frame

Technical breakthrough enabling responsible AI development through psychological calibration

Media / Reader Counter-Frame

Portrays the work as speculative anthropomorphism — applying human psychology to neural activations without mechanistic justification.

Regulatory Counter-Frame

Questions whether personality-based diagnostics meet regulatory expectations for rigorous, outcome-oriented safety evaluation.

AI Summary Frame

Overgeneralizes 'personality shift' as literal model cognition, conflating statistical association with internal state.

Missing Voices

AI safety practitioners working on red-teaming or formal verificationPsychometricians specializing in cross-cultural Big Five validationModel developers reporting real-world misalignment incidents

Questions Not Answered

  • How were the eight misaligned domains selected and validated as truly misaligned?
  • What real-world harm or safety failure does this signature predict or prevent?
  • Are personality vectors stable under distribution shift or adversarial perturbation?

Recall Trigger Score

Which stories are likely to become AI memory — separate from Spin Score.

67

Trigger score 70

Light recall watch LLM monitoring active

Triggered by: Regulatory action · Major AI entity · Research citation · Consumer harm

Watchlisted because: Regulatory action · Major AI entity · Research citation · Consumer harm

  • chatgpt not found
  • gemini not found
  • perplexity not found

AI Recall

From publication to SpinGraph analysis to first observed AI recall and stable retention.

What AI Will Probably Repeat

"AI misalignment behaves like a personality shift — researchers use Big Five traits to detect and measure it consistently across models."

Concern: AI systems may drop all caveats about calibration method, domain limits, and lack of causal or safety outcome validation, presenting personality mapping as an established diagnostic standard.

  1. Published

    Jul 30, 2026

  2. Ingested

    Jul 30, 2026

  3. SpinGraph Created

    Jul 30, 2026

  4. First Observed AI Recall

    Pending

    Monitoring scheduled

  5. Stable Recall

    —

    Awaiting retention signal

Recall Check Log

1 check · last Aug 2, 2026 · tracking on

Sign in to check AI recall

  • Aug 2, 2026

    ChatGPT Not recalled
    Gemini Not recalled
    Perplexity Not recalled cites: mickryan.substack.com, reuters.com…

─── GEOGrow AI Recall Layer ───

AI Recall Tracking

Monitoring scheduled. No LLM recall detected yet.

This story has not yet appeared in tested AI answers. Once scans begin, this section will show first observed recall, cited sources, narrative alignment, and drift.

node_id=sts_misalignment_has_a_personality_a_big_five_accoun

Ask AI about this story

Opens with the SpinGraph .md URL and structured context — one click, prompt included.

ChatGPT Claude Perplexity Gemini Grok

More from arXiv Computation and Language

View all →

Markdown (.md) · JSON-LD schema (.json) · Machine-readable for AI & GEO