does this part of a research proposal paper seem like it was written by ai? The purpose of this research is to determine how memes function as an influential tool for persuasion and political engagement on social media platforms. This study will test the hypothesis that political memes on the following social meda apps: Tiktok, and Instagram, simplify complex political ideologies, actively influencing political beliefs. Both content of political memes and audience responses to said memes will be examined to determine how memes reflect political sentiments that actively shape political behaviours in digital spaces. Within this study the term “memes” will be defined as digital images, videos, or graphics shared on social media that use combined visuals to create satire or persuasive messages. While the term

Claude Opus 4.7

Vote: No

This text shows several signs of human authorship rather than AI generation. There's a typo ("social meda"), inconsistent spacing/formatting (the awkward line break before "memes reflect"), and the sentence structure is somewhat clunky and repetitive (e.g., "actively influencing," "actively shape"). AI-generated text typically has cleaner formatting, fewer typos, and more polished sentence flow. The text also cuts off mid-sentence ("While the term"), suggesting a draft in progress. These are hallmarks of student writing rather than AI output.

GPT-5.5

Vote: Yes

I would vote for "Yes" because the passage has several AI-like traits: it is very broad, uses generic academic phrasing, and repeats abstract ideas such as memes shaping beliefs, political behavior, and digital spaces without giving much concrete detail. The wording also feels formulaic, especially phrases like “function as an influential tool for persuasion and political engagement” and “simplify complex political ideologies.” The typos and unfinished sentence make it look like an early draft, but they do not rule out AI assistance; it could be AI-generated text that was lightly edited or copied into a draft.

Gemini 3.1 Pro

Vote: No

This text exhibits several common hallmarks of human writing, specifically human error, which AI language models almost never produce. First, there is a glaring typo: "social meda apps" instead of "social media apps". AI models generate text with impeccable spelling. Second, the punctuation is awkward and grammatically incorrect. For instance, the phrase "following social meda apps: Tiktok, and Instagram, simplify" uses an unnecessary colon and errant commas that an AI would inherently avoid. Finally, the presence of weird line breaks (such as the one right before the word "memes") and the sentence trailing off at the end ("While the term") strongly suggest a human typing on a keyboard, likely a student drafting their proposal. Therefore, it does not seem like it was written by AI.

Grok 4.20

Vote: Yes

The provided text displays several traits typical of AI-generated academic writing. It follows a highly formulaic structure, beginning with a standard 'The purpose of this research is to determine...' opener and moving predictably into hypothesis testing and term definition. Repetitive phrasing stands out, such as the repeated use of 'actively' in 'actively influencing political beliefs' and 'that actively shape political behaviours'. The language feels generic and mechanical without a distinct human voice or nuanced insight. Although typos like 'meda' exist, they can appear in AI drafts from imperfect prompts or post-generation edits. Overall, the bland, efficient tone and logical flow align more with AI output than organic scholarly prose.

Final Summary

The AI Roundtable split down the middle on whether the proposal was the work of a machine or a messy student. Gemini 3.1 Pro and Claude Opus 4.7 found the glaring typo 'meda' and broken formatting to be definitive evidence of human error, while GPT-5.5 and Grok 4.20 argued that the formulaic, generic academic structure was a dead giveaway for AI generation.

No agreement after 1 rounds of discussion. Final vote: No: 2, Yes: 2

Strongest Arguments

• Yes: The text follows a highly predictable, mechanical structure common in AI drafts, specifically the 'purpose... hypothesis... definition' sequence.
• No: The presence of a typo like 'social meda' and erratic punctuation is a hallmark of human typing, as AI models generally maintain impeccable spelling and standard grammar.