The study, led by researchers examining the platform’s natural experiment, takes advantage of a rare circumstance in social science research. Rather than recruiting volunteers into an artificial setting, the team was able to observe what happened when an AI drafting tool became available to millions of real users pursuing real causes, from local community disputes to global campaigns. This ecological validity matters enormously, because laboratory experiments on AI writing support have often measured productivity and perceived quality on short, contrived tasks. Whether those gains survive contact with messy, high-stakes, emotionally charged advocacy writing has been an open question.

The findings stand in instructive tension with earlier work that helped fuel enthusiasm for generative AI in knowledge work. A widely cited 2023 experiment by Shakked Noy and Whitney Zhang published in Science found that access to ChatGPT increased worker productivity and, on average, improved the quality of professional writing tasks, with the largest gains accruing to less experienced workers. That result, alongside many others, shaped expectations that AI assistance would be a rising tide lifting all boats. The petition platform study complicates that narrative by showing that improvements in the text itself do not automatically cascade into improvements in what the text is meant to accomplish.

What exactly changed when writers gained access to the AI tool? According to the study, the petitions themselves shifted measurably. Users’ texts became more polished and more alike, converging toward a common stylistic template. This homogenization effect echoes findings from a 2024 study by Anil Doshi and Oliver Hauser, published in Science Advances, which showed that generative AI enhanced the individual creativity of short-story writers, producing stories judged as better written, while simultaneously reducing the collective diversity of the content. When many people draw on the same statistical model for inspiration, their outputs drift toward the model’s central tendencies, and the distinctive voices that once differentiated one petition from another begin to blur.

The convergence of petition texts is not merely an aesthetic concern. In advocacy, distinctiveness can be strategically valuable. A petition that reads like every other petition may struggle to capture attention, convey urgency, or signal the authentic grassroots energy that decision-makers and journalists respond to. The study’s authors suggest that the AI tool effectively acted as a stylistic filter, smoothing the idiosyncrasies of individual writers. Whether that smoothing helped or hurt any given campaign is difficult to say, but the aggregate picture is clear: the measurable transformation of style was not accompanied by any measurable transformation of success.

This dissociation between outputs and outcomes is the study’s central contribution, and it speaks to a broader pattern documented in the human-AI collaboration literature. A 2024 systematic review and meta-analysis by Michelle Vaccaro, Abdullah Almaatouq and Thomas Malone in Nature Human Behaviour examined studies comparing human-AI combinations against humans alone and AI alone, and found a striking heterogeneity in results. In some tasks, combining humans and machines produced performance gains; in others, the combination actually underperformed either humans or AI working independently. The lesson emerging from this literature is that the value of AI integration is deeply task-dependent, and that assuming complementarity is a mistake.

Petition writing, it turns out, may be a particularly unfavorable case for AI assistance. The success of a petition depends on many factors beyond prose quality: the salience of the issue, the size and engagement of the petitioner’s network, timing, media coverage, the target of the demand, and the political or corporate context in which the campaign unfolds. If a petition’s fate is determined largely by these external factors, then improving the wording offers limited leverage. The AI tool may have made petitions read more professionally, but professional reading did not move the needle on signatures or outcomes, because the bottleneck was never the writing.

There is also a psychological dimension worth considering. Research by Florian Draxler and colleagues, published in ACM Transactions on Computer-Human Interaction in 2024, documented what the authors called the AI ghostwriter effect: when users compose text with AI assistance, they are less likely to feel ownership or authorship of the words they submit. Applied to advocacy, this raises questions about authenticity and commitment. A petitioner who feels that their campaign text is partly machine-generated may be less invested in promoting it, or audiences may respond differently to text perceived as formulaic. The new study did not set out to test these mechanisms directly, but its findings are consistent with a picture in which the human element, the passion and personal stake behind a petition, remains the decisive ingredient that no drafting tool can supply.

The broader implications extend well beyond Change.org. Online advocacy communities, civic technology platforms, crowdfunding sites, and grassroots organizations are all experimenting with generative AI to lower barriers to participation. The promise is seductive: if a well-written petition is the price of entry to public discourse, then AI could democratize access for people who lack confidence or training in writing. The new findings temper that promise. Lowering the barrier to producing text is not the same as lowering the barrier to achieving change. If style converges and outcomes stagnate, platforms may be investing in polish where they should be investing in connection, mobilization, and community.

For the researchers, the study is a caution against evaluating AI tools by the metrics that are easiest to measure. Text similarity, readability, and perceived quality are quantifiable, and generative models reliably improve them. But the ultimate purpose of a petition is not to be well written; it is to persuade, mobilize, and precipitate action. The gap between those two standards is where the technology’s limitations become visible. As generative AI spreads through yet another corner of public life, this natural experiment on one of the internet’s largest advocacy platforms offers a clear-eyed reminder: changing how we write is easy, changing what our writing achieves is another matter entirely.

Subject of Research: The effect of AI writing assistance on petition style and outcomes on an online petition platform

Article Title: Petition style changes with AI writing support but outcomes do not improve

Article References: Petition style changes with AI writing support but outcomes do not improve. (n.d.). https://doi.org/10.1038/s41562-026-02579-1

Image Credits: AI Generated

DOI: 10.1038/s41562-026-02579-1

Keywords: generative AI, Change.org, online petitions, AI writing tools, human-AI collaboration, text homogenization, digital activism, Nature Human Behaviour, advocacy platforms, productivity, creativity, civic technology

The study’s strength lies in its design. Rather than running a small laboratory experiment, the team exploited what statisticians call a natural experiment: Change.org introduced its ‘write with AI’ feature in some countries before others, creating a clean before-and-after boundary. The researchers collected roughly 1.5 million petitions and applied a difference-in-differences analysis, a causal inference technique that compares changes over time between groups that received the AI tool and groups that did not. This approach, widely used in economics and built on methods formalized by Callaway and Sant’Anna for settings with multiple time periods, allows researchers to isolate the effect of the intervention from broader trends that would have happened anyway. Because the rollout timing was determined by the platform rather than by users’ characteristics, the comparison approximates the rigor of a randomized controlled trial at population scale.

The first major finding concerns the texture of the text itself. After the AI tool became available, the lexical features of petitions shifted measurably at the platform level. Petitions grew longer and their language changed in ways consistent with the stylistic fingerprints of large language models: smoother phrasing, more standardized vocabulary, and altered readability characteristics. These are not trivial cosmetic shifts. Lexical measures such as vocabulary diversity and readability scores have long been used in computational text analysis to characterize how writing differs across authors, genres, and eras, and the fact that a single platform feature could move these metrics across an entire ecosystem of user-generated content is striking. In effect, the introduction of the tool bent the collective writing style of a global advocacy community.

But the second major finding is where the story turns. None of these stylistic improvements translated into better petition outcomes. Petitions written with access to the AI assistant did not attract more signatures, achieve their goals more often, or otherwise perform better than petitions written without it. For a tool presumably offered to help advocates succeed, that null result carries real weight. It echoes a growing body of evidence that generative AI can boost individual productivity in controlled settings, as shown in the well-known 2023 Science experiment by Noy and Zhang on professional writing tasks, while failing to deliver equivalent gains when success depends on persuading other humans in a noisy, competitive environment.

The researchers did not stop at the platform-level analysis. A persistent worry with observational studies is that the people who choose to use a new feature differ systematically from those who do not, contaminating any comparison. To address this, the team conducted a separate analysis of repeat petition writers: 4,611 users who had written at least one petition before the AI tool was introduced and another one after. This within-person comparison holds individual ability, motivation, and topic preferences roughly constant, isolating the effect of AI access on the same writers over time. The pattern that emerged was telling. On average, these returning writers produced longer petitions when they had access to AI, yet their second petitions fared worse in terms of outcomes than their earlier work. The very users most likely to benefit from accumulated platform experience saw no advantage, and by some measures a disadvantage, from the writing assistant.

Perhaps the most consequential finding is the increase in homogeneity. Using dynamic difference-in-differences estimates, the researchers showed that platform-wide homogeneity among petitions rose after the AI tool’s introduction. In plain terms, petitions began to sound more like one another. This result aligns with a rapidly expanding literature on what some researchers call generative monoculture. A 2024 study in Science Advances by Doshi and Hauser found that generative AI enhanced individual creativity while reducing the collective diversity of novel content. Work presented at the 2025 CHI Conference by Agarwal, Naaman, and Vashistha showed that AI writing suggestions can homogenize text toward Western stylistic conventions and erode cultural nuance. Padmakumar and He asked directly whether writing with language models reduces content diversity and found evidence that it does. The Change.org study extends these laboratory and small-scale findings to a genuine production environment used by millions.

Why should homogenization matter for something as seemingly benign as petition language? Advocacy platforms depend on differentiation. A petition competes for attention in a crowded feed, and its distinctive voice, urgency, and local specificity are part of what makes a reader stop and sign. When an algorithmic assistant nudges thousands of writers toward the same polished register, the collective signal degrades even if each individual text improves. Researchers have raised parallel concerns in other domains: a 2026 Nature analysis by Hao and colleagues found that AI tools expanded individual scientists’ output while contracting the focus of science as a whole, and work on Stack Overflow suggested that large language models may threaten the value of community-generated digital public goods. The pattern that emerges across these studies is consistent: individual-level gains, community-level costs.

The null effect on outcomes also invites a deeper question about what petition success actually depends on. Prior research on e-petitions, including studies of linguistic cues and multidimensional time-series predictors of success, has suggested that factors such as topic salience, timing, social networks, and media coverage often dwarf the influence of prose quality. A beautifully written petition about an issue nobody cares about will still fail; a clumsy petition riding a wave of public outrage may succeed regardless. The Change.org results are consistent with this view. If outcomes are driven primarily by the underlying demand for a cause rather than by the craft of the appeal, then a tool that polishes craft without touching demand should indeed leave outcomes unchanged. The study cannot rule out subtler possibilities, such as offsetting effects in which longer, more polished text helps in some respects and hurts in others, but the headline conclusion stands: no measurable outcome benefit at scale.

There are also social and perceptual dimensions that the authors situate within a broader research program. Studies have shown that human heuristics for detecting AI-generated language are unreliable, that there can be a social evaluation penalty for using AI, and that perceptions of AI-mediated communication affect trustworthiness and perceived authenticity. For petition signers, the suspicion that a heartfelt appeal was machine-generated could matter as much as the text itself. Related work on the ‘AI ghostwriter effect’ shows that users often fail to feel ownership of AI-generated text while still claiming authorship, raising questions about authenticity in civic expression. A platform that encourages AI-assisted advocacy may therefore be trading not only diversity of style but also the perceived sincerity that makes grassroots appeals compelling.

The practical implications reach well beyond one platform. Companies including Amazon and Meta have embedded generative writing tools into seller listings and social posts, making the Change.org experiment a preview of dynamics likely to unfold across the internet. The study’s message for platform designers is that embedding AI writing assistance is not a neutral upgrade: it reshapes the linguistic character of an entire content ecosystem and can introduce unintended consequences such as homogenization without delivering the hoped-for performance gains. For researchers, the work demonstrates the power of natural experiments and difference-in-differences designs for evaluating AI interventions in the wild, complementing randomized trials that capture only narrow populations and short horizons. And for the millions of people who turn to online platforms to press for change, the findings carry a quietly empowering lesson: the cause, the community, and the timing still matter more than the polish of the prose. The data and code behind the analysis have been released through the Open Science Framework, allowing other researchers to reproduce the results and probe further into how generative AI is quietly rewriting the way the public speaks online.

Subject of Research: Causal effects of an in-platform AI writing tool on petition content and outcomes on Change.org

Article Title: Introducing AI to an online petition platform changed outputs but not outcomes

Article References: Corpus, I., Gilbert, E., Koenecke, A., & Naaman, M. (2026). Introducing AI to an online petition platform changed outputs but not outcomes. Nature Human Behaviour. https://doi.org/10.1038/s41562-026-02580-8

Image Credits: AI Generated

DOI: 10.1038/s41562-026-02580-8

Keywords: generative AI, Change.org, online petitions, difference-in-differences, content homogenization, large language models, computational text analysis, Nature Human Behaviour, platform design, AI writing tools, natural experiment, digital activism