or instance,models like ChatGPT discovers statistical patterns inenormous quantities of online text, which may containmistakes, biases, and obsolete data

Ifauthors utilize these tools to aid in writing or manuscriptpreparation, they must take accountability for the accuracy ofthe generated content

In June 2022, a French preprint server named HALpublished a fourth essay co-written by an earlier chatbot by thename of GPT-3 and was later accepted for publishing in apeer-reviewed journal [11] .

They may fabricate citations andreferences when producing academic papers, promptingexperts to conclude that they cannot be relied upon to givecorrect information or credible references.

Asscientists, our paramount apprehension is that these AIlanguage bots are ill-equipped to assimilate novel information,conjure up revelations, or engage in profound scrutiny,ultimately curtailing the scope of scholarly discourse.

As a result, it is critical for researchers, particularly ethicists,to debate the trade-off between using AI to accelerateknowledge creation and the possible loss of human capabilityand control in the research process. Creativity, originality,education, training, and human relationships will almostcertainly remain important in performing relevant and novelresearch.

If we only emphasize performance and efficiency, the role andimpact of individuals may decline as AI technology advances.

Although AI tools excel atrepeating established knowledge, they fall short in recognizingand producing novel findings. In fact, they may even strugglein determining if a new discovery is genuine, aberrant, orgroundbreaking.

It is critical toassess and improve the validity of language models so thatresearchers may utilize the technology in an effective andethical manner.

This form of conversational AI integration in research andpublication has the potential to change the discipline,presenting both advantages and drawbacks. It can acceleratethe innovation process, shorten the time it takes to publish, andenhance equality and diversity in scientific viewpoints byassisting people who struggle with writing [7]

The rise of AI automation might also result in the loss ofunique writing styles, replacing them with a uniform style.This raises questions about whether the homogenization ofwriting is a step towards greater understanding or a loss ofdistinctive features

Notably, ChatGPT is being used byresearchers for a variety of applications, including composingessays, summarizing literature, improving papers, detectingresearch gaps, and producing computer code and statisticalstudies [2] It has also been used in academic contexts togenerate research papers and graphic features such as figuresand tables and is more often seen in such publications [3] .

As of now, thewidespread adoption of ChatGPT is inevitable, but if usedrecklessly and without proper oversight, it poses a threat to theacademic publishing industry. To prevent this, it is necessaryto take a more thoughtful and diligent approach to modeltraining and invest in AI output detection systems. WhileChatGPT has the potential to revolutionize the field, we stillneed to prepare to use it effectively and comprehensively.

Eventually, it will beimpossible to detect their use in written assessments.Academia, and in turn PME, must accept this realityand determine new ways to evaluate a student’s indi-vidual development that uphold academic integritywhile also still building agile and adaptive leaders whocan think critically for themselves in stressful environ-ments.

“AI or die” is the latest mantra being tossed aroundin technology circles.

Students must still be able to communicate in writ-ing; however, this skill can be developed and assessedvia shorter, timed on-the-spot written assessments.

AI-generated essays may contain citations that areeither fabricated or incorrectly applied.

Only a human can take their yearsof real-world experience and translate it into a deepassessment of doctrinal concepts. The absence of thisdepth of scrutiny is indicative of AI-generated text.

Sometimesit colligates several sources into one “ghost” citation.

Additionally, it does not gauge the accuracy or legit-imacy of the source, which often results in suspectcitations to say the least.

colligates

incredulous

AI writing systems are trained to produce plau-sible-sounding text but can introduce errors with157MILITARY REVIEW September-October 2025AI-ASSISTED WRITINGauthoritative confidence.

This directly links to number nine, “Misalignmentwith Prompt Details,” unguided AI essays will oftencontain “fluff ” unrelated to the topic, especially when aprompt demands a specific page or word count. While(AI image by author)September-October 2025 MILITARY REVIEW158instructors are used to seeing this in human-draftedwork, AI “fluff ” will be confidently written and will trickthe reader regarding its relevance.

However, note that repeated prompts can elevatesome of the other indicators such as ghost citations, ashift in vocabulary, and the dearth of a human voice.

Listening to their paper via the MicrosoftWord “Read Aloud” function can help identify a changein tone.

Second, robust written essays with future deadlinesmust be replaced by shorter written assignments andoral boards to truly test a student’s comprehension.

A “debrief ” can occurwhere the student is asked a series of pointed questionsabout their essay. If the student genuinely developed thepiece, they then should be able to defend and describetheir thought process in assembling the work.

Paragraphs will often follow a rigid structurethat feels mechanical and lacks the dynamic organizationtypically found in well-developed student writing.

An author’s unique way of expressing themselveswill manifest in their tone, syntax, and choice of words.If this suddenly disappears in a completed essay, iteither means they had a strong proofreader or AI wrotethe piece.

AI-generated writing often appears grammaticallyflawless yet devoid of the individual perspective, per-sonal, or professional voice expected of a human being.

An instructor must take note of their students’writing style when viewing these presubmissions.Coupled with other writing samples such as emails,journal entries, or discussion posts, it will give a solidindication of a student’s writing ability. If the finalessay differs significantly from these samples, this isan indication that the text is AI generated.

Notembracing the possibilities found within AI would be amonumental folly on our part.

Clear and professional written communication isessential for conveying ideas, presenting information, persuading stakeholders, and maintainingpositive business relationships

explicate

Just as speakers transmit emotion through voice, writers can transmit a range of attitudes through writing, from excited and humorous to somber and critical. These emotions create connections among the audience, the author, and the subject, ultimately building a relationship between the audience and the text.

If you want to brag about a good grade, you may write the post to please family members. If you want to describe a funny moment, you may write with your friends’ senses of humor in mind. Even at work, you send e-mails with an awareness of an unintended receiver who could intercept the message.