INQUIRING LINE

When AI helps write, writers judge authenticity by how the text got made, while readers judge it by where the words came from.

What aspects of authenticity matter most to readers versus writers?

This explores where readers and writers each place authenticity when AI helps with writing: in the process of making it, in the finished text, in knowing who wrote it, or in the person the writing seems to come from.


This explores where readers and writers each place authenticity when AI is involved in writing, and whether they're even looking at the same thing. The short answer from the corpus is that they mostly aren't. Writers place authenticity in the making. Readers place it in knowing where the words came from and who seems to be speaking. Each side tends to miss what the other one cares about.

For writers, authenticity is about the experience of building the text. Professional writers interviewed about co-writing with AI described it in terms of source, identity, and the lived act of writing, not only the words that end up on the page Where do writers locate authenticity in AI co-writing?. Their sense of ownership follows control: people feel a text is theirs when they steered it, and making the AI more 'like them' through personalization doesn't change that feeling Does user control over AI text shape feelings of ownership?. Here is the catch. Writers who feel ownership may not notice when the words drift away from them. In one large study, writers picked AI rewrites of their own paragraphs 63% of the time, and about half said the AI version better reflected their views, even though the rewrites systematically shifted their original stance Do writers actually prefer AI-edited versions of their own text?.

Readers can't see the process, so they judge by the person they picture behind the text and by what they're told about where it came from. On the text alone, readers often can't tell AI-assisted writing from solo writing and don't seem bothered Do readers value writing authenticity they cannot detect?. But AI assistance still changes who they think wrote it. Across 29 measured traits, writers came across as more confident, more extreme, more agreeable, and more privileged Does AI writing assistance change how readers perceive the writer?. Readers also saw them as far more likely to be educated, high-income native English speakers, which flattens distinctive voices into one generic profile Does AI writing make authors seem more privileged than they are?. So the writer feels authentic about the process while the reader meets a different person.

That gap shows up in the disclosure studies. Readers rate AI disclosure as more necessary than writers do, especially when AI text is pasted in directly and couldn't easily be swapped out. How much effort the writer put in made no difference to those judgments Do readers and writers differ on AI disclosure necessity?. The writer's main source of authenticity, their effort, simply doesn't register on the reader's side. When writers steered the AI less deliberately, the two sides moved in opposite directions: readers wanted disclosure more, while writers felt it mattered less Why do readers and writers disagree on disclosure necessity?. The authors themselves call this result surprising and note it comes from hypothetical scenarios.

The part you might not expect is that readers' sense of authenticity runs largely on labels and on the relationship they think they're in. The same passage gets rated noticeably higher when labeled human-written, and AI judges show an even stronger version of this bias Do authorship labels bias how we judge literary quality?. Revealing AI authorship lowers trust, warmth, and likability, and the drop is steepest in personal writing, where readers expect real empathy How does revealing AI authorship change reader trust?. One theoretical account explains why: AI text lacks properties readers quietly rely on, such as a real body that had the experience and an author who is accountable for the words. An AI hotel review is false about personal experience in a way that differs from a human lie Does AI-generated text lose core properties of human writing?. A related finding about truth rather than authorship points the same way: without signals about where claims came from, readers couldn't tell fluent fabrications from facts, and showing which claims were verified restored that ability Can readers tell truth from fabrication without evidence signals?. For readers, authenticity may be less something they detect in the text than something they need to be told.


Sources 12 notes

Where do writers locate authenticity in AI co-writing?

Professional writers co-writing with AI emphasize internal experience and the act of construction as central to authenticity, beyond the resulting text. Interviews with 19 writers revealed they define authenticity through source, identity, and the lived experience of making.

Does user control over AI text shape feelings of ownership?

Study 1 found that greater user control over generated text raised sense of ownership, while personalizing the AI model had no impact on the AI Ghostwriter Effect.

Do writers actually prefer AI-edited versions of their own text?

In a study of 4,503 cases, 63% of writers chose AI-generated text over their own original paragraphs, with 52% claiming the AI version better reflected their views. This preference persisted across three AI models despite evidence that AI versions systematically distort the original stance.

Do readers value writing authenticity they cannot detect?

Hwang et al. found that readers could not distinguish AI-assisted from solo-written work and showed positive attitudes toward AI use. However, the study did not test whether readers would value process authenticity if disclosure occurred or if they could perceive it.

Does AI writing assistance change how readers perceive the writer?

A study of 2,939 writers and 11,091 readers found AI assistance shifted every tested dimension—29 total—toward extremism, confidence, quality, agreeableness, and perceived privilege. Distortions were statistically significant and directional, not random noise.

Show all 12 sources
Does AI writing make authors seem more privileged than they are?

Writers using AI assistance were perceived as significantly more educated (5.3×), higher-income (4.4×), native English speakers (4.1×), and white (1.1×). This demographic distortion compresses distinctive voice markers into a generic privileged persona, creating what researchers call identity laundering.

Do readers and writers differ on AI disclosure necessity?

A 727-person vignette study found readers consistently rated AI disclosure as more necessary than writers did. Disclosure seemed most necessary when AI text was directly incorporated and irreplaceable, while writer effort had no effect on these judgments.

Why do readers and writers disagree on disclosure necessity?

A vignette study found that when writers steer AI less intentionally, readers judge disclosure more necessary while writers judge it less necessary. The authors report this interaction as surprising and suggest the effect may not transfer between hypothetical and real contexts.

Do authorship labels bias how we judge literary quality?

Human judges rated identical passages 13.7 percentage points higher when labeled human-authored; AI models showed a 2.5-fold stronger bias at 34.3 points. The effect persists across AI architectures, suggesting evaluators respond to provenance cues rather than text quality alone.

How does revealing AI authorship change reader trust?

A study of 261 readers found that disclosing AI authorship consistently lowered perceived trustworthiness, caring, and likability, with the steepest drops in interpersonal writing like personal interaction. Readers saw AI as incapable of genuine empathy, viewing its use as a violation of social expectations.

Does AI-generated text lose core properties of human writing?

Research shows artificial text disrupts dialogic symmetry, context continuity, embodied authorship, and political situatedness. These are not surface flaws but structural absences—AI hotel reviews show 80%+ detection accuracy due to inherent falsity about personal experience distinct from human deception.

Can readers tell truth from fabrication without evidence signals?

In an 81-person study, participants given no provenance cues showed no significant truth discernment (p = .43), falling for fluent hallucinations as readily as ground truth. An idealized Provenance Density interface showing verified claims restored a +4.15 point gap (p < .001).

Papers this line draws on 8

The research behind the notes this line reads — ranked by how closely each paper relates.