AI Essay Writer – The future of essay writing?

Introduction

In recent years there has been an explosion in AI writing apps with uses expanding into all areas of modern life. One particular niche that has seen a huge increase in AI usage is education, with AI essay writers now being capable of writing academic essays and assignment on any subject.

What is an AI Essay Writer

The definition of an AI Essay Writer is an application which uses AI (Artificial Intelligence) to generatively create essays, assignments and other academic papers using machine learning algorithms. A typical AI essay writer will use one or more text inputs (prompts), to create a written essay which may include tables, lists, headings, in-text references and citations, and other elements common to academic papers.

Key differences exist between the latest generation of AI essay writing applications compared to those available pre-2021. With the launch of OpenAI’s ChatGPT, AI writing transformed from what was previously simple “lookup and copy” methods of creating essays, to new algorithms which can “generate unique content” based on research and data available in ChatGPT’s own extensive data stores, user input, and via integrations with 3^rd party suppliers (plugins). We call this new generation of AI writing applications “Generative AI”.

With so many AI essay writer applications to choose from, it can be a dizzying choice to select the right application for you, or to even know if you should be using a generative AI writer at all.

In the next section we cover the basic pro’s and cons of using AI essay writing apps, and consider the potential impacts you might see on your studies. From there we will go on to talk about the 3 best-known AI essay writing tools available right now.

Advantages of Using an AI Essay Writer App

For most students there are a bewildering array of options available, but most AI essay writer apps will offer at least the following advantages:

1. Increased speed of writing – AI apps can create generative content within a few minutes, compared to what could be several hours work to write the essay manually.
2. AI writers have Broad abilities to research topics and access related material from academic journals and other publications.
3. Almost all AI writers have the ability to translate content between multiple languages with relative ease and accuracy
4. Using AI to write an essay means that you don’t need to have the knowledge that would normally be required to write the paper yourself.
5. AI essay writers can provide proofing and referencing for any work generated, saving yet more time.

Disadvantages of Using an AI Essay Writer

1. Your university will almost certainly know that the essay was not written by you.
2. Although generative essays can appear well written on first view, it is often the case that research is not up-to-date or complete.
3. Choices of sources used for citations may not be relevant to your particular course or level of study – again leading to your lecturer knowing you did not write the essay yourself.
4. Although AI has advanced greatly, it still fails to create credible reasoned arguments or reflections, and lacks critical thinking.
5. Writing styles from even the best AI writers have patterns that AI detection apps can often recognise.

As you can see above, there are many reasons to consider using an AI writer, but for students and especially students in Higher Education the risks of using AI generated essays are incredibly high. If you are caught handing in a paper that is recognised by your university as being AI generated you risk failing your course, or even expulsion from your academic institution.

Which is the Best AI Essay Writer?

There are currently three main contenders for the best AI writing app:

1. ChatGPT 4
2. Google Bard
3. Claude.ai

Each of the AI writers above has the ability to create essays, in varying levels of detail and accuracy.

ChatGPT

Website: https://chat.openai.com/

Probably the best known of all the AI writing apps, ChatGPT 4 offers users a highly accomplished generative writing app that is capable of creating academic essays and other papers. ChatGPT uses “tokens” to handle user input, and output generated test. These tokens are limited to 10,000, which may sound like a lot but the more complicated the user input, the less tokens are available to generate the essay. This means that ChatGPT is best for content of less than 750 words (approximately)

The platform understands academic styles of citation and can create reasonably well written content that could be suitable for high school studies or similar. The quality of content and the level of writing tend not to be suitable for higher level topics, or subjects requiring reasoning or critical thinking.

ChatGPT is capable of writing in multiple languages and can also work with code such as Python and Javascript.

Google Bard

Website: https://bard.google.com/chat

This AI writing app is provided by Google, the well-known search engine. Although Bard is less well known than ChatGPT it is just as capable, and has the ability to write essays to a reasonable academic standard, although not enough to be able to achieve a pass grade for most subjects.

Bard has slightly better abilities to work with tables and data than it’s rival ChatGPT, although is nowhere near as accomplished in this area as Claude.ai. With access to more recent information than ChatGPT, the Google Bard writer offers students a great tool for carrying out research

One major advantage of Bard is its ability to work with “Spoken Responses”, meaning it can read out-loud the generated essay, and help you with pronunciation of any unfamiliar words it used.

Claude.ai

Website: https://claude.ai

Claude is currently the gold standard in AI essay writing, and the quality of outputted content is substantially better than ChatGPT 4 or Bard. This is because Claude uses 10,000 tokens shared between input and output, making it possible to create 5,000 word essays with a single prompt, and because of the amount of tokens available, Claude can write to required word counts with relative ease.

Claude also has another killer advantage over both Bard and ChatGPT; users can upload documents into the AI system which can then be used to perform various tasks. A good example is how a university marking rubric can be uploaded and used to grade or mark work. Claude can accept any kind of file upload including Word, Excel and PDFs, and then analyse the content and draw conclusions and analysis from the data within.

However, despite the obvious power of Claude.ai, the generated essays are still not up to a standard that would be seen from a human writer. Much like Bard and ChatGPT, Claude fails down when it comes to it’s choice of sources, access to current information, lack of ability to reason or reflect, and a style of writing that is easily recognised as being AI.

Claude has definite uses in education, especially in areas like suggesting potential essay topics and titles, creating essay plans and structures, and checking completed work against a rubric or content guide.

Should I use an AI essay writer?

At the time of writing (October 2023) we cannot recommend using AI to write essays for anything more complicated than high school assignments. Although all 3 of the AI apps above can create in-depth and structurally well-referenced essays, the content lacks human flair, is often inaccurate and can easily be detected by both AI detection apps, and lecturers who are familiar with the student.

Using an AI essay writer can be a huge risk, and penalties for using such apps can be extremely severe with students being thrown off their courses, often without any refund of course fees.

-Barclay Littlewood, Owner, UKEssays.com

However, if you want to research potential topics, get help creating a structure for your essay, or are looking for ideas for potential sources, AI essay writers offer some real benefits.

Example AI Essay Writer Content

As a simple test, we used the same prompt to ask the three AI apps to write an answer for an academic question:

Question / Prompt: In less than 500 words, explain the Gibbs reflective cycle. The explanation should be written in British English and be appropriate for a 1^st year undergraduate degree at an English institution. Provide authoritative references to accompany the explanation, using the APA style of academic citation.

---

Example output from ChatGPT 4:

Gibbs Reflective Cycle

Gibbs Reflective Cycle, developed by Graham Gibbs in 1988, is a structured model used for reflecting on experiences, particularly within the context of professional practice. This model encourages individuals to think systematically about the phases of an experience or activity, and it is widely used in various professional fields, including nursing, education, and business.

The cycle consists of six stages:

1. Description: At this stage, you describe the event or situation in detail, answering questions like: What happened? Who was involved? Where and when did it take place? This is a factual account without any interpretation or judgement.
2. Feelings: Here, you reflect on your feelings and thoughts during the event. How did you feel before, during, and after the situation? It's essential to be honest and open about your emotions, as this helps in understanding your reactions.
3. Evaluation: This stage involves assessing the experience. What went well? What didn't go as planned? Were the outcomes positive or negative? It's about analysing the situation and your role in it.
4. Analysis: Delve deeper into the experience to understand the underlying reasons for your actions and the outcome. You might consider theories, literature, or other knowledge to make sense of the situation. This stage is crucial for learning as it helps to understand the broader context.
5. Conclusion: Reflect on what you could have done differently. What have you learnt from the experience? What would you change if faced with a similar situation again? This stage is about recognising areas for personal or professional growth.
6. Action Plan: Based on your reflections, decide on the steps you will take if the situation arises again. This might involve seeking further training, changing your approach, or seeking advice. The aim is to ensure better outcomes in the future.

Using the Gibbs Reflective Cycle can lead to deeper learning and improved practice. By systematically working through each stage, individuals can gain insights into their actions and behaviours, enabling them to develop and grow in their professional roles.

Reference: Gibbs, G. (1988). Learning by doing: A guide to teaching and learning methods. Further Education Unit. Oxford Polytechnic: Oxford.

---

Example output from Google Bard:

The Gibbs reflective cycle is a framework for structured reflection developed by Graham Gibbs in 1988. It is a cyclical model of reflection, meaning that it can be used repeatedly to learn from different experiences. The cycle consists of six stages:

1. Description: What happened?
2. Feelings: What were your thoughts and feelings during and after the experience?
3. Evaluation: What went well and what could be improved?
4. Analysis: Why did things happen the way they did? What factors contributed to the outcome?
5. Conclusion: What have you learned from this experience?
6. Action plan: What will you do differently in the future based on what you have learned?

The Gibbs reflective cycle can be used in a variety of settings, including education, healthcare, and social work. It is a valuable tool for learning from experiences, both positive and negative. It can also help to identify areas for improvement and develop strategies for change.

Example of using the Gibbs reflective cycle

A first-year undergraduate student is reflecting on a presentation they gave in their English class.

Description: The student gave a 10-minute presentation on the history of the novel. They had prepared carefully and felt confident about their material. However, when they started to give their presentation, they felt nervous and their voice trembled. They also made a few mistakes in their delivery.

Feelings: The student felt disappointed in themselves after the presentation. They were embarrassed by their mistakes and worried that their classmates had judged them negatively.

Evaluation: The student evaluated their presentation and identified the following areas for improvement:

• They needed to practice their delivery more beforehand to reduce their nervousness.
• They needed to be more careful about proofreading their slides to avoid making mistakes.

Analysis: The student analyzed why they felt nervous during their presentation. They realized that they had a fear of public speaking. They also realized that they had put a lot of pressure on themselves to give a perfect presentation.

Conclusion: The student learned that it is important to be prepared for presentations and to practice their delivery beforehand. They also learned that it is important to be realistic about their expectations and not to put too much pressure on themselves.

Action plan: The student decided to join a public speaking club to help them overcome their fear of public speaking. They also decided to start practicing their presentations in front of a mirror.

Authoritative references

The following are authoritative references on the Gibbs reflective cycle:

• Gibbs, G. (1988). Learning by doing: A guide to teaching and learning methods. Further Education Unit, Oxford Polytechnic.
• Dewey, J. (1933). How we think: A restatement of the relation of reflective thinking to the educative process. D.C. Heath.
• Schon, D. A. (1983). The reflective practitioner: How professionals think in action. Basic Books.
• Boud, D., Keogh, R., & Walker, D. (1985). Reflection: Turning experience into learning. Routledge.

Conclusion

The Gibbs reflective cycle is a valuable tool for learning from experiences. It can be used in a variety of settings to improve professional practice and personal development.

---

Example Output from Claude AI

Here is a 473 word explanation of the Gibbs reflective cycle in British English with APA references:

The Gibbs reflective cycle is a theoretical model introduced by Graham Gibbs in 1988 to provide a structured approach for reflecting on experiences. It offers a framework for examining experiences in order to engage in the process of continued learning. The cycle involves six key stages:

Description The first stage involves describing the experience in detail. This includes recalling what happened during the experience such as what you did, what others did, what the context was, and what your reactions were. The aim is to develop a comprehensive account of the experience.

Feelings
The second stage involves evaluating the feelings you experienced during the event. This includes exploring both positive and negative emotional reactions to the experience. The goal is to identify how you felt about the experience at the time.

Evaluation
The third stage involves analysing the experience through evaluating it. This means making judgments about what was good and bad about the experience. It also involves considering what went well and what did not. The aim is to develop an understanding of the overall impact of the experience.

Analysis
The fourth stage involves breaking the experience down into component parts to try to understand it better. This means exploring the reasons behind your feelings and reactions. The goal is to gain insight into the causes influencing your experience.

Conclusion
The fifth stage involves summarising what you have learned from reflecting on the experience. This includes drawing conclusions about what you might do the same or differently in the future. The aim is to develop self-knowledge from the analysis of the reflective process.

Action Plan
The final stage involves outlining the proposed actions that you intend to take as a result of the reflection. This should identify the specific steps you will implement based on the insights gained from all the previous stages. The goal is to use the reflection to achieve real change (Gibbs, 1988).

Overall, the Gibbs reflective cycle provides a step-by-step approach for reflecting on experiences with the ultimate goal of learning from them and applying any lessons to future situations. It offers a valuable model for purposeful reflection.

References
Gibbs, G. (1988). Learning by doing: A guide to teaching and learning methods. Oxford Polytechnic: Oxford.

---

Conclusion

As can be seen in the example generated essay content above, each AI essay writer has it's own unique way of approaching the prompt, and there is a huge difference in the output from each app.

For this relatively simple task there was a clear winner in terms of content depth, Google Bard, with a detailed explanation of the topic and a wide selection of sources referenced. 

However, in terms of following the prompt, and providing a clear explanation of the Gibbs Reflective Cycle, the content generated by Claude much more clearly defines what the cycle is, and process and aims for each individual step. The content from Claude.ai may appear more simplistic than the explanation generated by Google Bard, but the actual substance of the text is far more meaningful. Also, for a 500 word description, a single authoritiative reference makes more sense than the 4 references provided by Bard. It's also iportant to note that Claude.ai cited these sources in-text; (Gibbs, 1988), whereas Bard only created a list of references with no in-text citations, essentially rendering the references as meaningless.

Overall, the AI Essay Writer that most competently answered the question posed in the prompt was Claude.ai, and tests performed on longer prompts and responses also lean towards Claude.ai being the market leader for generative AI in the education sector.

That said, based on the overall quality of content, the level of analysis and understanding of the topic, none of the AI writers would achieve a pass grade in a UK university setting, and to the trained eye, all 3 examples above are clearly written by AI, and not a student.