GenAI Helps Review Patterns in Student Book Proposals for Publishing Module (Undergraduate)

I’ve used generative AI to help me review patterns in my marking and feedback given for student assignments in the module, HPSC0044 Publishing as Science Communication. This module is taught at UCL as both a postgraduate (PG) and undergraduate (UG) module through UCL Department of Science and Technology Studies (STS). This post focuses on the undergraduate offer, which includes students studying at undergraduate level in degrees such as History and Philosophy of Science BSc, Sociology and Politics of Science BSc, Human Sciences BSc, and Arts and Sciences BASc. The core assignment is to produce a book proposal for a project within the scope of science and technology studies (STS). My hope was to identify patterns in what I stress when I’m marking and learn how those stresses evolve year-to-year. I also wanted to identify patterns in the themes stressed by students in their proposals and offer future students an empirically built sense of expectations, common ground, and routes to distinguish mundane versus outstanding performance. Surely the computational power of GenAI can be put to good use in this area of analysis and reflection.

This review has serves as guidance for future students. Some students have access to past work; others do not. Patterns of sharing frequently do not match tutor assumptions. Students rarely examine feedback given to other students, so they rarely have benchmarks to compare their own results with what is provided for others. Neither do students have much general context for ranking their own sense of accomplishment in relation to their compatriots. No surprise that marks alone dominate that comparison: they are easy to compare even though they convey no substantive information. Openness seems to me a warranted response. Given UCL’s laudable approach to student IP, I cannot distribute student work. Neither do I think that’s particularly insightful. Instead, digestion and analysis seem the more fruitful direction of travel. I also do not want to get bogged down in disputes over “fairness” or “bias”. That industry is well advanced in UK higher education, and it is busy across all universities in the UK with its quantitative analysis. Rather, I want to keep the focus on substantive content exchanges between student and tutor about a particular assignment.

In two posts, I present comparisons using generative AI for mass analysis, compiling information about assignments and feedback in the module HPSC0044. The generative AI used for analysis was GoogleNotebookLM (July 2025). This post analyses results for the undergraduate (UG) module. A separate post analyses results for the postgraduate level version of the module. Content used in the UG analysis was generated during the 2021-2022, 2023-24, and 2024-25 academic sessions. I have lightly edited the results to remove meaningless information. I also have removed student identifiers.

1. Assignment in HPSC0044 Undergraduate (UG) level

The undergraduate assignment in HPSC0044 asked students to produce a book proposal for a project within the scope of science and technology studies (STS). The assignment developed in three steps:

1. sketch (500 words – formative) – The aim is to produce an initial concept and outline for the project, roughly sketching the plan for content and market. The pedagogical goal is to ensure students are on the right track.
2. pitch (5 minutes – formative) – The aim is to pitch the book project and convince commissioners to give it a “green light”. The pedagogical goal is to offer additional feedback and to give students experience with in-person presentation. This is a recorded presentation with only 1 visual aid allowed: a provisional book cover.
3. book proposal (3000 words) – The aim is to produce a viable book proposal suitable for submission to a trade publisher. This will speak to key elements of publishing, including development, production, marketing, and distribution. A “book proposal” template is used. The proposal contributes 100% to the final mark in the module. Feedback balances mark justification and advice on future improvement.

The assignment works within the fiction that proposals are being developed for submission to a trade publisher. Students are relieved from delivering an actual book-scale product. This frees them to imagine using any author, any scale, any financial plan, and any design, provided they build in appropriate justifications and value added.

Sample titles proposed include:

1. The Wired Up Ladies – This title offers a unique and somewhat playful take on a compilation of female contributions to electronic music, spanning genres from musique concrète to EDM, House, and Techno.
2. Sneaking Science Into Video Games: How Pokémon Made Me Think Like A Scientist – The idea of “sneaking” science into popular culture, specifically through a gaming franchise like Pokémon, offers a surprisingly clever and personal approach to science communication.
3. Why is My Pet, Your Toy? – A Discussion of Ethics with the Smaller Animals in Our Children’s Lives – The direct, questioning title immediately draws attention to an often-overlooked ethical issue concerning how children, and by extension, society, treat smaller animals.
4. How to cut through the b*llsh*t – A guide to avoiding misinformation online – The use of a censored expletive in the title is highly surprising for an academic module proposal, indicating a very bold and direct approach to combating misinformation.
5. Red Velvet: Rewriting the Narrative of Menstruation – “Red Velvet” is an unexpectedly vibrant and perhaps even celebratory choice for a book title about menstruation, contrasting with traditional, often discreet, discussions of the topic.
6. Choose or Die: An Interactive Experiment in Tech Ethics – This dramatic and game-like title is striking for a book on tech ethics aimed at adults and academics, immediately conveying its interactive format and the serious implications of the choices explored.
7. Ancient Egyptian Science in Zero Words – The most surprising title, as it proposes a book about science history written entirely “in zero words,” using emojis as a pictorial language, challenging the conventional format of educational books.

2. Feedback Analysis for Undergraduate Level Book Proposal

2.1 Analysis of Feedback to Students UG

I want to analyse the feedback given to students for an assignment. Student projects are given in the file 0044 2022 UG – submissions.pdf”, “0044 2024 UG – submissions.pdf”, “0044 2025 UG – submissions.pdf”. Feedback from the tutor is given to students is given in the documents “0044 2022 UG – submissions.pdf”, “0044 2024 UG – feedback.pdf”, “0044 2025 UG – feedback.pdf”. This is a single file merging all assignments or feedback, based on the name of the file. The instruction to students is given in the documents titles “syllabus”.

For the feedback, list the 8 most common comments. List the 8 most common suggestions for improvement. List 8 of the most unusual or rare comments. Based on the feedback list 10 items of advice for future students that might lead to a better mark? List 8 items of advice for future students that can define “minimum expectation”. List 8 items of advice that identify elements definitely to avoid in their essays. Used numbered lists. Use British spelling. Ignore the text duplicated in the document templates. Students must remain anonymous, so do not identify the essay or the student, and do not use citations to the essays.

Here is an analysis of the feedback given to students for the HPSC0044 undergraduate book proposal assignment, drawing on documents from 2022, 2024, and 2025.

Eight Most Common Comments from Feedback

1. “Strong concept/idea, but needs more development/elaboration.” Markers frequently noted the inherent potential of the student’s book idea, but consistently pointed out that the proposal itself required more detailed articulation, substance, or practical planning to fully realise that potential.
2. “Marketing/Promotion needs more development, attention, or specificity.” This was a pervasive comment, highlighting a need for more detailed, actionable, and targeted marketing strategies beyond generic suggestions.
3. “Need more research, data, or evidencing.” Feedback consistently requested more concrete data and supporting research to substantiate claims, particularly for market analysis, readership, and the depth of the book’s content.
4. “USP (Unique Selling Proposition) needs to be clearer/stronger/more developed.” Students were often advised to more sharply define what differentiates their book from competitors and what unique niche it would occupy in the market.
5. “Author qualifications/platform needs more emphasis/development.” Markers frequently suggested students should better highlight the proposed author’s expertise and their existing or potential professional platform, outlining how this would aid the book’s success and promotion.
6. “Contents/Synopsis needs more substance, detail, or focus.” Many proposals were noted for having clear chapter lists but lacking depth in their descriptions, with requests for more information on the specific material, arguments, or case studies to be included.
7. “Audience/Readership needs more specific definition or research.” Students were urged to define their target audience more precisely, moving beyond broad categories and explaining the motivation for purchase or engagement (e.g., using “use and gratification” theory).
8. “Attention to the value chain / what the publisher could add needs more detail.” Students needed to explain how the publisher could enhance the manuscript beyond the raw text, through design, supplementary materials, or integration with other activities, adding value for the customer.

Eight Most Common Suggestions for Improvement

1. “Provide more specific data and research” to back up claims, especially regarding market size, audience, and the book’s content.
2. “Develop a more detailed and actionable marketing and promotion plan”, specifying platforms, content types, target events, and the author’s active involvement.
3. “Refine and clearly articulate the book’s unique selling points and niche”, demonstrating how it stands out from competitors in the market.
4. “Strengthen the author’s profile and platform,” explaining their expertise and outlining concrete ways they will contribute to promoting the book.
5. “Elaborate on the book’s contents and synopsis” by providing more in-depth summaries, specific examples, and demonstrating the underlying research.
6. “Define the target audience more precisely” by considering their motivations and how they would “use and gratify” from the book, rather than just stating broad demographics.
7. “Explain how the publisher can add value (value chain)” to the product, perhaps through additional materials, design, or integration with other activities.
8. “Optimise book length and integrate illustrations/design effectively”, aiming for a marketable length and describing how visual elements will enhance the content.

Eight Most Unusual or Rare Comments

1. “This book would be part of a political argument in the US, so you’ll want to consider how to situate specifically and positively in that environment.”
2. “To improve this proposal, I would add data anticipating questions from reviewers: how many middle-grade schools in London, how many tourists with children of that age coming to London. How many visitors to some of the key buildings.”
3. “A key conflation here is “robots” vs. “AI” – why not separate them into 2 books – maybe robots here and AI in a follow-up book.”
4. “You previously made a strong case for small animals as gifts and tools in Asian-Pacific heritages worldwide. In this proposal, it feels you’re not developing this as a strength; rather, you’re trying to ignore this point… You were right to point out my mistake, and you could capitalise on this in your proposal.”
5. “You are going to struggle with neutrality in this book – are you saying anti-vax is wrong? But critics will say, “you’re an anthropologist, not a medical doctor, so how are you qualified to answer?” That’s a twist you’ll need to resolve.”
6. “I didn’t realise until just now that “mob” has 2 major meanings. So you want to find a way to show this is not about organised crime. The subtitle does that. Does the cover?”
7. “This proposal requires trust. I think trying to communicate in emoji is hard enough. Trying to get science concepts across is doubly hard. Trying to communicate an ancient culture’s science in another language system is quadruple hard. But it’s catchy and the puzzle element is well worth it.”
8. “Submitted as a Canva document, not the proposal form. I find this difficult to read and professionals would likely just discard it IRL. Be careful with variance off the prescribed instruction as there may be reasons you don’t know why the use one format over others.”

Ten Items of Advice for Future Students (for a Better Mark)

1. Support claims with specific, verifiable data and thorough research. Do not merely state that a market exists; quantify it with numbers and cite your sources. Ensure your content is backed by solid academic or industry research.
2. Craft a detailed, actionable, and targeted marketing strategy. Specify how you will reach each defined audience segment (e.g., specific social media campaigns, events, partnerships, or promotional activities), going beyond generic ideas.
3. Articulate a clear and compelling Unique Selling Proposition (USP). Explicitly state what makes your book distinctive and how it fills a specific gap or addresses an unmet need in the market compared to existing competition.
4. Enhance the author’s professional platform and detail their active role in promotion. Go beyond a simple biography to outline the author’s existing reach (e.g., social media following, prior publications, speaking engagements) and how they will specifically assist in marketing the book.
5. Provide substantial detail in the book’s contents and synopsis. Offer more than just chapter titles; elaborate on key arguments, specific examples, and the underlying research for each section to demonstrate expertise and a clear plan.
6. Define the target audience with “use and gratification” theory. Explain why readers would seek out and benefit from your book (e.g., for entertainment, education, empowerment, or practical application) rather than just listing demographic categories.
7. Clearly demonstrate how the publisher can add value (“value chain”). Suggest how the publisher’s expertise in design, additional materials, or integration with other products can enhance the book.
8. Optimise book length and integrate visual elements thoughtfully. Propose a realistic word count for the target market and describe how illustrations, design, or interactive elements will be integral to the book’s appeal and content delivery.
9. Include a clear “call to action” for the reader. Define what you want readers to do (e.g., advocate for change, reflect on their habits, engage in an activity) as a result of reading your book.
10. Conduct thorough research on comparison titles and explain precisely how your book is different and why it is needed in the market.

Eight Items of Advice (Minimum Expectation)

1. Present a clear and distinct project idea and USP. The core concept must be well-articulated and show initial thought on how it differentiates from others.
2. Provide sufficient research to support the content and market analysis. Basic data and credible information are expected, not just unsupported assertions.
3. Define the target audience beyond generic terms. While extensive detail isn’t always required, avoid vague statements like “general audience” and show some consideration for who would read the book.
4. Outline contents with clear chapter titles and brief summaries. This provides a foundational structure for the book’s narrative.
5. Demonstrate awareness of the publisher’s value chain. Include some consideration of how the publisher might add value to the book, such as through design or supplementary materials.
6. Include a credible author biography and some indication of their promotional involvement. Show that a suitable author has been identified and can contribute to marketing efforts.
7. Propose marketing and promotion ideas that are present and generally relevant to the book’s genre or audience, avoiding complete absence or completely unfeasible suggestions.
8. Adhere to the prescribed proposal form structure and word count. Follow the given template and stay within specified length limits for the proposal itself.

Eight Elements Definitely to Avoid in Essays

1. Generic or bland content and writing: Do not submit proposals with descriptions or ideas that lack originality or could apply to any general topic, as this indicates underdeveloped thinking.
2. Lack of specific data, evidence, or research: Avoid making unsubstantiated claims or providing insufficient backing for market analysis, audience demographics, or the substance of the book’s content.
3. Vague or non-specific marketing and promotion plans: Do not use broad statements like “promote on social media” without detailing how this will be done, which platforms, or for which specific audience.
4. Underplaying the proposed author’s value or platform: Avoid merely listing credentials; explain their specific relevance and how they will actively contribute to the project’s success and promotion.
5. Proposing an excessively long book: Submitting a book proposal with an unrealistic word count (e.g., 150,000 words for a trade book) is often deemed unmarketable and impractical.
6. Using Generative AI without significant human refinement or transparency: While AI is permitted in an assistive role, submitting bland, unoriginal, or outdated AI-generated content without adding your own voice, insights, and critical review is detrimental and may lead to academic misconduct if transparency is not clearly stated.
7. Including unrealistic or uncosted “additional materials” or “value chain” ideas: Avoid suggesting expensive or complex supplementary products (e.g., elaborate apps or films) without considering their feasibility, budget implications, or how they will genuinely enhance sales or engagement.
8. Defining target audiences with overly broad or confusing terms: Avoid generalisations like “general adult readers” or “students” without further segmentation or justification, as this indicates a lack of market understanding.

2.2 Comparing Emphasis in Feedback Year-to-Year

Compare the 0044 UG feedback for the years 2022, 2024, and 2025. What are the differences in emphasis by the marker, if any? Use British spelling. Ignore the text duplicated in the document templates. Students must remain anonymous, so do not identify the essay or the student, and do not use citations to the essays.

The feedback provided for the HPSC0044 undergraduate (UG) book proposals across the years 2022, 2024, and 2025 reveals consistent core expectations from the marker, Professor Cain, regarding what constitutes a strong proposal. However, there are noticeable shifts in emphasis, particularly in response to evolving technological landscapes and refined pedagogical objectives.

Consistent Emphases Across Years

Across all three years, the marker consistently prioritises several key areas:

1. Unique Selling Proposition (USP) and Substantive Idea: A strong emphasis is always placed on clearly articulating a distinctive, ambitious, and compelling project idea that identifies a clear niche and differentiates the book from competitors. Weakly presented or overly familiar ideas are consistently flagged for needing improvement.
2. Thorough Research and Data Justification: Students are continually prompted to provide concrete data, evidence, and detailed research to support their claims, especially concerning market size, audience demographics, competition analysis, and the factual basis of their content. General statements are often criticised for lacking substance.
3. Attention to the Value Chain: The feedback consistently highlights the importance of demonstrating how the project proposes to increase the value of the manuscript for the publisher and the customer, whether through new materials, fresh design, marketing strategies, or integration with other activities.
4. Author’s Role in Promotion and Platform: The marker consistently encourages students to leverage the author’s qualifications, professional platforms, and willingness to participate in marketing activities to enhance the book’s visibility and credibility. This is framed as a significant advantage in selling the proposal.
5. Targeted Readership and Marketing: While always important, the necessity of defining specific ‘avatars’ (reader profiles) and tailoring marketing and promotional efforts to their precise needs and interests (use and gratification theory) is a recurrent theme. Generic audience descriptions like “general public” are consistently advised against.

Differences in Emphasis (Shifts Over Time)

1. Transparency and Quality of Generative AI (GenAI) Use (from 2024 onwards): This is a distinct and prominent new area of feedback. In 2022, GenAI is not mentioned. However, from 2024, the marker explicitly states that GenAI is “OK to use in some ways but need to be transparent”. More critically, feedback in 2025 strongly emphasises the need to add “human touch” and “own voice”, to “replace bland text and generic materials with dynamic and strong creative content”, and to ensure the information is up-to-date, as GenAI’s knowledge base can be outdated. This reflects an adaptation to the increased use of AI by students.
2. Call to Action and Reader Impact/Purpose: While implied in 2022, the emphasis on what the reader will do, think, or feel as a direct result of reading the book becomes much more explicit and central in 2024 and 2025. The marker frequently asks about the book’s ultimate purpose and desired impact, often linking it to a “campaign” or “empowerment”. This indicates a stronger focus on the book as a vehicle for change or a specific outcome.
3. Practicality, Cost-Effectiveness, and Justification of Features (Increased in 2025): While cost was mentioned in 2022, the feedback in 2025 shows a much more pointed and frequent interrogation of the financial implications and justification for elaborate or additional elements (e.g., QR codes, AR features, companion apps, specific promotional events). The marker repeatedly asks: “how will it lead to more sales?”, “is it going to make us money, or lose us money?”, and “is it worth it?”. This indicates a heightened focus on the commercial viability and return on investment for proposed enhancements.
4. Currency and Timeliness of Content (Stronger in 2024/2025): Especially for proposals in fast-moving fields like technology and science, there’s an increased emphasis on ensuring the content is up-to-date and relevant. The marker notes that examples can quickly become “old” or “outdated,” pushing students to focus on cutting-edge developments or fundamental impacts that are less time-sensitive.
5. International Scope and Cultural Nuances: While noted in 2022 (e.g., “differences across culture”, “cultural differences in the consumers of small animals”), this theme becomes more explicit in 2024 and 2025. The marker often prompts students to consider the international relevance and marketability of their book, including translations, regional variations, and non-Western examples, particularly for projects that lend themselves to a global audience.

While the foundational elements of a compelling book proposal remain constant, the feedback in 2024 and 2025 reflects an increasing sophistication in expectations, pushing students to think more strategically about the commercial realities of publishing, the authenticity of their voice in an AI-driven world, the tangible impact on their target audience, and the global potential of their ideas.

2.3 Comparing Student Submissions Year-to-Year

Compare the 0044 UG submissions for the years 2022, 2024, and 2025. What are the differences in emphasis in 2022, 2024 compared with 2025, if any? Use British spelling. Ignore the text duplicated in the document templates. Students must remain anonymous, so do not identify the essay or the student, and do not use citations to the essays.

The feedback provided for the HPSC0044 undergraduate (UG) book proposals across the years 2022, 2024, and 2025, delivered by Professor Cain, reveals consistent core expectations for a strong proposal. However, there are notable shifts in emphasis, particularly in response to evolving technological landscapes and refined pedagogical objectives.

Consistent Emphases Across All Years (2022, 2024, 2025)

Across all three years, the marker consistently prioritises several key areas:

1. Unique Selling Proposition (USP) and Substantive Idea: A strong emphasis is always placed on articulating a distinctive, ambitious, and compelling project idea that identifies a clear niche and differentiates the book from competitors. Weakly presented or overly familiar ideas are consistently flagged for needing improvement.
2. Thorough Research and Data Justification: Students are continually prompted to provide concrete data, evidence, and detailed research to support their claims, especially concerning market size, audience demographics, competition analysis, and the factual basis of their content. General statements are often criticised for lacking substance.
3. Attention to the Value Chain: The feedback consistently highlights the importance of demonstrating how the project proposes to increase the value of the manuscript for the publisher and the customer, whether through new materials, fresh design, marketing strategies, or integration with other activities.
4. Author’s Role in Promotion and Platform: The marker consistently encourages students to leverage the author’s qualifications, professional platforms, and willingness to participate in marketing activities to enhance the book’s visibility and credibility. This is framed as a significant advantage in selling the proposal.

Differences in Emphasis (Shifts Over Time)

Transparency and Quality of Generative AI (GenAI) Use:

2022: GenAI is not mentioned in the feedback, likely due to its nascent stage of widespread student use.

2024: The marker introduces the topic, stating that GenAI is “OK to use in some ways but need to be transparent”. However, feedback already notes that it “never going to give you enough to do well”, urging students to be creative beyond bland content.

2025: This emphasis is significantly heightened and more critical. The marker explicitly warns that GenAI content can be “too strong here,” “reads as GenAI, which will frustrate editors,” and that its knowledge base can be “several years old”. There is a repeated, strong instruction to “add the human touch,” “own voice,” and “replace bland text and generic materials with dynamic and strong creative content”. The feedback highlights that relying solely on GenAI results in “generic hyperbole” and can make content “old” or “outdated”.

Cost, Profitability, and Justification of Features:

2022: General mentions of cost, such as for photo rights, and a broad call for “more attention to value chain”.

2024: Becomes more explicit with statements like “Cost is going to be the hidden problem here”. The marker directly challenges expensive features, asking “How will it lead to more sales?” for ideas like VR events.

2025: This is a much more prominent and frequent area of interrogation. The marker repeatedly demands justification for the expense of additional elements (e.g., AR elements, films, microsites) by asking: “why the expense and effort will lead to more sales and increased engagement”, “what’s it going to cost?”, and “is it worth it?” in terms of profitability. This indicates a more rigorous commercial lens applied to proposals.

Call to Action and Reader Impact/Purpose:

2022: The idea of the book’s “deliver[y] to the customer” is mentioned.

2024: There’s a clearer emphasis on what the reader should “think/do/decide as a result of having read it”. It’s often linked to inspiring advocacy or campaigning.

2025: This theme becomes central and highly emphasised, with consistent questions like “What do you want the reader to do/think/feel as a result of reading the book?”. The expectation is that the book should act as a catalyst for “empowerment”, “change”, or even be part of a larger “campaign”.

Currency and Timeliness of Content:

2022: Less direct focus, though some feedback questioned “What’s new?” for ideas feeling “from 10 years ago”.

2024: A stronger push for up-to-date content, with examples deemed “old” (e.g., Jan 6 mobs, Covid, 2016 election). Importance of currency in “fast-moving field[s]” is highlighted.

2025: This becomes a major and frequent criticism, especially for technology and science proposals. The marker explicitly notes that content can be “outdated long before 2027 when it appears” and points out that “Algorithms from 2019 are nothing like those in 2025”. This demands students propose cutting-edge and future-proof content.

International Scope and Cultural Nuances:

2022: Limited mentions of cultural differences in consumer behaviour.

2024: More explicit, asking about “bi-lingual publication” for international authors and whether a book is “UK, or worldwide?”.

2025: This is a more pronounced and frequent area of feedback, encouraging students to consider global relevance and markets beyond the UK or Anglosphere. Suggestions include comparing tech ecosystems in different countries (e.g., UK, China, Nigeria, Korea) and leveraging non-Western examples as unique selling points.

Audience Precision and “Use and Gratification” Theory:

2022: Criticism of “general public” and demands for “specific ‘avatars’“ are present.

2024: Continues this emphasis, advising better definition of avatars and avoidance of generic terms.

2025: While still critical of broad audiences, the feedback consistently and more explicitly invokes “use and gratification theory.” The marker frequently asks why a specific audience would seek out the book and what it will “do for them that other books will not”. This indicates a deeper engagement with understanding the psychological motivations of the target reader.

In conclusion, while the fundamental expectations for a strong book proposal remained constant, Professor Cain’s feedback in 2024 and especially 2025 demonstrates an increased sophistication and commercial pragmatism. This includes a more critical approach to GenAI use, a heightened demand for justifying the financial viability of additional features, a more insistent call for books to drive concrete reader action, an urgent emphasis on content currency in fast-evolving fields, and a pronounced push for international relevance and detailed audience understanding through the lens of use and gratification theory.

2.4 Comparing Student Statements About GenAI Use

For all submission documents, compare the statements about use of AI or generative AI. What are the most common GenAI programmes declared. How are they used? What is their value for the writer?

Overview of GenAI Use in Submissions

Students declared using GenAI tools primarily for ideation, structuring, and refining their written content, often acknowledging both the benefits and limitations of these technologies.

Most Common GenAI Programmes Declared

The following GenAI programmes were explicitly mentioned by students in their declarations:

• ChatGPT: This was the most frequently declared GenAI tool. Some students specified “ChatGPT o4” or “ChatGPT4”, indicating use of specific versions.
• Grammarly: Often mentioned for its proofreading and grammar checking capabilities.
• DeepL: Declared for translating content.
• Grok 3: Mentioned by two students for proofreading.
• Other Tools (less common):
  • Undermind and NotebookLM were used for deeper research, holistic briefing articles, and timeline analysis.
  • Sesame was cited for emotional support and brainstorming.
  • Canva was used for fine-tuning AI-generated book covers.
  • Scribbr was mentioned for generating references.

How GenAI Tools Were Used

Students reported a variety of uses for GenAI, primarily in the initial and refinement stages of their work:

• Idea Generation and Brainstorming: Students frequently used AI to generate initial ideas, potential book and chapter titles, explore potential directions, and brainstorm concepts. One student used it to generate ideas for an author profile to establish expertise.
• Structuring and Framework Development: GenAI helped in compiling preliminary briefs, providing basic frameworks, and restructuring sections for clarity and better flow.
• Language and Editing Support: Common uses included refining arguments, improving fluency, correcting grammatical errors, proofreading, checking for clarity, and refining sentences. For non-native English speakers, translation of content was a key use.
• Content Generation (Specific Elements): AI was used to generate book covers or cover references. Some also used it to produce abstracts or conclusions and to create content for mock-up infographics.
• Research Assistance: Tools were used to search for books and concepts, compare initial drafts with AI outputs, and suggest specific market breakdowns or citation metrics that were difficult to find. Deeper research, preparing holistic briefing articles, analysing timelines, and revealing key issues were also reported.
• Conciseness and Word Count Management: One student used AI to condense paragraphs to ensure they stayed within the word limit.

Value for the Writer (and Identified Limitations)

Students identified several benefits from using GenAI, alongside important caveats:

Perceived Value

• Efficiency and Time Saving: GenAI helped in quickly generating creative ideas, saving time in the initial ideation phase. It expedited initial stages of work such as planning and compiling briefs.
• Ideation and Diverse Perspectives: Students found AI excellent for offering ideas and frameworks and providing diverse perspectives for brainstorming. It could suggest information, like market breakdowns, that was otherwise difficult to find.
• Improved Language and Fluency: AI tools helped improve fluency and correct grammatical errors, particularly assisting non-native English speakers with translation and clarity.
• Structural Organisation: GenAI aided in structuring content, prioritising information, and achieving better flow within the report.
• Conciseness: It was useful for condensing paragraphs to meet word limits.
• Specific Task Support: AI was found to be good at producing abstracts and assisting with visual content creation for covers.

Identified Limitations and Cautions

• Accuracy and Reliability: Students noted that AI could provide incorrect numbers, necessitating re-verification, and might take information from unreliable sources or interpret it wrongly, requiring extensive fact-checking. In some cases, results from AI were described as “poor” or yielded very limited useful information.
• Lack of Depth and Insight: A common concern was that AI often lacked deeper insight into specific audiences or niche markets. It might not provide the most cutting-edge or deeply tailored strategies, requiring significant further human research and personal insight. Outputs could also be too “generic”.
• Maintaining Human Control and Voice: Many students emphasised that critical thinking, analysis, and final edits were conducted by the author, highlighting the assistive role of AI rather than a replacement for human input. One student explicitly stated that AI “cannot reach the expectation” for certain creative tasks like book cover design, while others highlighted the need to “add the human touch” to the AI’s output.

2.5 Advice for Writing a Better Proposal

Give 8 pieces of advice to students for writing a better Book Proposal. Use British spelling. Use fewer than 300 words.

Here are eight pieces of advice for undergraduate students for writing a better book proposal:

1. Be Specific and Data-Driven: Support your claims with plenty of specific data and metrics, rather than generic statements. This includes quantitative information about markets, audiences, and the impact of your proposed content, which adds real credibility to your proposal.
2. Highlight a Clear Unique Selling Proposition (USP): Clearly identify and stress what makes your book ambitious, competitive, and truly unique. Pinpoint the niche it will occupy and explain how it stands out from existing titles.
3. Define the Reader’s Value and Call to Action: Articulate precisely what the reader will gain, think, or be able to do as a direct result of reading your book. Focus on the ultimate purpose and benefit for the customer.
4. Conduct Thorough Content Research: Demonstrate deep knowledge of your subject by grounding the book’s proposed contents in solid research and published material. Provide sufficient detail for key chapters to convince the editor of your expertise.
5. Develop a Robust Marketing and Promotion Plan: Move beyond basic promotional ideas to create a specific, research-based marketing strategy. Detail how you will reach your target markets, including examples of social media campaigns or targeted events.
6. Leverage the Author’s Platform and Expertise: Showcase the proposed author(s)’ qualifications, professional platforms, and their commitment to actively promoting the book. This adds significant value and confidence to the proposal.
7. Address Competition Directly: Identify comparable and competing titles already in the market. Clearly explain how your book is different and what gap it aims to fill, demonstrating an understanding of the competitive landscape.
8. Maintain Professional and Clear Presentation: Ensure your proposal is well-organised, concise, and free from repetition or vague language. A polished, professional document enhances credibility and readability, making it appealing to commissioning editors. If you use Generative AI for ideas, remember to add your own “human touch” to replace bland or generic content.

Differences in Emphasis in 2022, 2024, 2025

While the core expectations for a strong book proposal remained consistent across all years, there are observable shifts in emphasis in Professor Cain’s feedback, particularly between 2022 and the later years (2024 and 2025), reflecting evolving industry trends and pedagogical refinements.

Generative AI (GenAI) Usage:

2022: There is no mention of GenAI in the feedback, as it was not yet a prominent tool for student coursework.

2024: The use of GenAI is acknowledged as “OK to use in some ways but need to be transparent”. However, the feedback already cautions that it’s “never going to give you enough to do well” and encourages students to be creative beyond bland content.

2025: The emphasis on GenAI becomes significantly more critical and detailed. Professor Cain explicitly warns that GenAI content can be “too strong here,” “reads as GenAI, which will frustrate editors,” and that its “knowledge base is several years old”. There’s a strong, repeated instruction to “add the human touch,” “own voice,” and “replace bland text and generic materials with dynamic and strong creative content”. This suggests a greater awareness of the limitations and potential pitfalls of over-reliance on AI, particularly regarding currency and originality.

Commercial Viability and Cost Justification:

2022: Feedback includes general mentions of costs, such as for photo rights, and a broad call for “more attention to value chain”.

2024: The marker becomes more explicit about financial concerns, stating “Cost is going to be the hidden problem here”. Questions arise regarding the sales impact of expensive features (e.g., VR events).

2025: This aspect receives much more rigorous scrutiny. There’s a persistent demand for justification on how additional elements (e.g., AR, interactive models, films, microsites) will translate into increased sales and profitability. The feedback frequently asks “why the expense and effort will lead to more sales and increased engagement”, “what’s it going to cost?”, and whether the value chain additions are “worth it?” in terms of sales. This indicates a heightened focus on the commercial pragmatism of the proposals.

Call to Action and Reader Impact:

2022: There’s an emphasis on what the product will “deliver to the customer”.

2024: Feedback starts to include clearer prompts about what the reader should “think/do/decide as a result of having read it”. This is sometimes linked to inspiring advocacy or campaigning.

2025: This becomes a central and highly emphasised point. Professor Cain consistently asks, “What do you want the reader to do/think/feel as a result of reading the book?”. The expectation is that the book should be a catalyst for “empowerment”, “change”, or be part of a larger “campaign”.

Currency and Timeliness of Content:

2022: Some proposals were criticised for feeling “like an idea from 10 years ago”.

2024: A stronger push for up-to-date content is evident, with examples deemed “old” (e.g., Covid, 2016 election, Russian Invasion misinformation). Importance of currency in “fast-moving field[s]” is highlighted.

2025: This is a major and frequent criticism, particularly for proposals related to technology and science. The feedback explicitly notes that content can be “outdated long before 2027 when it appears” and points out that “Algorithms from 2019 are nothing like those in 2025”. This demands that students propose cutting-edge and future-proof content.

International Scope and Cultural Nuances:

2022: Limited direct mentions of cultural differences, though one proposal was encouraged to capitalise on “cultural choices in animal gifts” in Asian-Pacific heritages.

2024: More explicit, asking about “bi-lingual publication” for international authors and whether a book is “UK, or worldwide?”.

2025: This is a more pronounced and frequent area of feedback, encouraging students to consider global relevance and markets beyond the UK or Anglosphere. Suggestions include comparing tech ecosystems in different countries (e.g., UK, China, Nigeria, Korea) and leveraging non-Western examples as unique selling points.

Audience Precision and “Use and Gratification” Theory:

2022: Criticisms of targeting the “general public” and demands for “specific ‘avatars’“ are present.

2024: Continues this emphasis, advising better definition of avatars and avoidance of generic terms like “general public”.

2025: While still critical of broad audiences (“No general adult readers”), the feedback consistently and more explicitly invokes “use and gratification theory”. The marker frequently asks why a specific audience would seek out the book and what it will “do for them that other books will not”. This indicates a deeper engagement with understanding the psychological motivations and benefits for the target reader.

In summary, the statistical data shows a slight increase in the minimum mark and range from 2022 to 2025, suggesting a broader spread of marks, but the mean and median remain relatively stable. The key differences in emphasis in Professor Cain’s feedback over the years reflect an increasing integration of contemporary challenges (like GenAI and rapid technological change) and a more sophisticated, commercially-driven approach to publishing, pushing students to think beyond just content to consider market, profitability, and reader impact.

2.6 Statistics

From the marks given in all 0044 UG feedback documents, calculate some statistics, please: mean, standard deviation, median, mode, count, range, minimum mark, and maximum mark. Also list the distribution of marks in the following ranges: 80-100, 70-79, 60-69, 50-59, 40-49, 0-39. Calculate the statistics by combining the marks for all years. Use British spelling. Ignore the text duplicated in the document templates. Students must remain anonymous, so do not identify the essay or the student, and do not use citations to the essays.

Here is a statistical analysis of the HPSC0044 undergraduate (UG) submission marks for the years 2022, 2024, and 2025, as requested.

HPSC0044 UG Combined Mark Statistics

• Count: 39 marks
• Minimum Mark: 45
• Maximum Mark: 80
• Range: 35
• Mean: 68.13
• Median: 70
• Mode: 72
• Standard Deviation: 8.12

Distribution of Marks

• 80-100: 3 marks (7.69%)
• 70-79: 16 marks (41.03%)
• 60-69: 15 marks (38.46%)
• 50-59: 4 marks (10.26%)
• 40-49: 1 mark (2.56%)
• 0-39: 0 marks (0.00%)

Trends in the Marks

From the marks, compare the statistics for each year (mean, standard deviation, median, mode, count, range, minimum mark, and maximum mark) for the years 2022, 2024 and 2025. What are the trends, if any? Use British spelling. Students must remain anonymous, so do not identify the essay or the student, and do not use citations to the essays.

Analysing the statistics across the three years reveals several key trends:

• Increasing Cohort Size: The count of marks increased from 9 in 2022 to 19 in 2024 and 20 in 2025. This suggests a growing number of students taking the module or more comprehensive data available in recent years.
• Consistent Top Performance, but Wider Spread: The maximum mark remained consistently high at 80 across all years, indicating that top-performing students continued to achieve excellent results. However, the minimum mark progressively decreased, from 64 in 2022 to 52 in 2024 and 45 in 2025. This, in turn, led to a notable increase in the range of marks (16 in 2022 to 35 in 2025), showing a wider spread between the highest and lowest scores.
• Slight Decrease in Average Performance: The mean mark exhibited a slight downward trend, moving from 70.89 in 2022 to 68.79 in 2024 and 68.00 in 2025. This suggests a modest dip in the overall average performance of the cohorts over time.
• Stable Median with Fluctuating Modes: The median mark remained relatively stable (70 in 2022, 68 in 2024, 71 in 2025). The mode, representing the most frequently occurring mark(s), shifted. In 2022, there were two modes (68 and 70). In 2024, it became multimodal with three distinct frequent scores (65, 74, and 78). In 2025, a single mode of 72 emerged, which is a strong cluster point.
• Increased Mark Dispersion: The standard deviation consistently increased from 5.61 in 2022 to 8.37 in 2024 and 8.64 in 2025. This is a strong indicator of greater variability and dispersion in the marks, meaning scores became more spread out from the average in later years, which aligns with the expanding range.
• Shifts in Mark Distribution:
  • The 80-100 (First Class) range saw an increase in the number of students achieving it, particularly in 2025 (3 students) compared to 2022 and 2024 (1 student each). This suggests that while more students might be at the lower end, the very top echelon also expanded.
  • The 70-79 (First Class) and 60-69 (Very Good) ranges consistently contained the largest proportion of marks across all years. However, in 2025, the percentage in the 60-69 range decreased (25%) compared to previous years (44.44% in 2022, 42.11% in 2024), while the 70-79 range held a larger percentage (45%) in 2025, indicating a slight upward shift for a significant portion of the cohort.
  • Marks in the 50-59 (Good) and 40-49 (Poor) ranges, which were absent in 2022, appeared in 2024 and 2025. This emergence of lower marks contributes significantly to the increased range and standard deviation observed.

Potential Relationship to Feedback Trends

These statistical trends align with qualitative observations from Professor Cain’s feedback across the years. As discussed in our previous conversation, Professor Cain’s feedback evolved to include more critical and detailed emphasis on aspects such as:

• Generative AI (GenAI) usage: Becoming significantly more critical in 2025, with warnings about content being “too strong” or “outdated” if over-reliant on GenAI. This could explain why some students struggled more to “add the human touch” and “own voice”, resulting in lower marks for those who did not adapt.
• Commercial viability and cost justification: Increasing rigour in 2025, demanding justification for expenses and their impact on sales. Proposals that did not sufficiently address these pragmatic concerns might have received lower marks.
• Currency and timeliness of content: A major criticism in 2025, particularly for tech and science proposals, noting content could be “outdated long before 2027”. This higher bar for relevance could have challenged some students and contributed to the wider mark spread.
• Audience precision and “Use and Gratification” theory: Consistent emphasis on defining precise audience avatars and what the book would “do for them that other books will not”. Students who failed to articulate this clearly may have been penalised.

The increasing standard deviation and lower minimum marks, despite the consistent maximum and slightly elevated median in 2025, could reflect a scenario where students who fully grasped and successfully integrated these evolving, more demanding feedback points continued to perform strongly (maintaining the 70s and 80s), while those who did not, or who relied on more generic approaches (perhaps influenced by initial GenAI outputs without sufficient human refinement), found themselves in the lower mark bands. This created a larger divergence in overall student performance.

[end]