{"id":1761590,"date":"2025-03-16T18:53:08","date_gmt":"2025-03-16T18:53:08","guid":{"rendered":"https:\/\/instatext.io\/?p=1761590"},"modified":"2025-04-01T16:53:12","modified_gmt":"2025-04-01T16:53:12","slug":"why-you-cant-risk-trusting-ai-detectors-for-academic-submissions","status":"publish","type":"post","link":"https:\/\/instatext.eu\/why-you-cant-risk-trusting-ai-detectors-for-academic-submissions\/","title":{"rendered":"Why you can\u2019t trust AI detectors for academic submissions"},"content":{"rendered":"\n

Do you want to know why you can’t risk relying on publicly available AI detectors<\/strong> for your academic submissions if you don’t want to ruin your chances of getting published and jeopardise your reputation<\/strong>, academic standing<\/strong>, and career prospects<\/strong>? Here is what you need to know. <\/p>\n\n\n\n

What are the concerns about AI-generated content?<\/h2>\n\n\n\n

Leading academic institutions, including the Committee on Publication Ethics (COPE<\/a>), ICMJE<\/a>, the Council of Science Editors (CSE<\/a>), MDPI<\/a>, Elsevier<\/a>, Cambridge<\/a>, and Springer<\/a>, state that generative AI tools do not meet the required criteria<\/strong> of authorship, originality, and confidentiality and that their output may contain biases or inaccuracies.<\/p>\n\n\n\n

What are AI detectors?<\/h2>\n\n\n\n

AI detectors emerged to help determine whether a piece of content has been created or modified using AI, as generative AI tools such as ChatGPT have made it more difficult to distinguish between human and AI-generated content. <\/p>\n\n\n\n

In the context of written content, AI detection tools, which are often built into writing and editing software, analyse the text in a similar way to spell checkers or plagiarism detectors and generate scores. These scores provide an estimated percentage of AI-generated content within a document. <\/p>\n\n\n\n

Are AI detectors reliable?<\/h2>\n\n\n\n

There is a growing list of tools that promise users they can detect content written by AI with the highest accuracy. However, studies have shown that AI detectors have<\/a> their limitations and are<\/a> \u201cneither accurate nor reliable<\/strong>,\u201d producing a high number of false positives and false negatives.\u00a0<\/p>\n\n\n\n

In a famous experiment<\/a>, SEO specialist Dianna Mason ran the Declaration of Independence through an AI content detector, which found that it was 98.51% AI-generated, even though it was written in 1776.\u00a0\u00a0<\/p>\n\n\n\n

Another challenge is to distinguish between fully AI-generated and AI-revised content, with the latter being even more difficult to recognise as it \u201ctends to share more characteristics with human-generated content,\u201d according to<\/a> a paper examining the detection of AI-generated application documents in higher education.<\/p>\n\n\n\n

\u201cA general and universally effective detection model would be extremely useful, but appears to be beyond the reach of current technology and detection methods,\u201d the authors note. <\/p>\n\n\n

\n
\"\"<\/figure><\/div>\n\n\n

Now that we have discussed that publicly available tools for detecting AI-generated content are not accurate and reliable, the question is: Are there other types of AI detectors that are more accurate and reliable but are not publicly available?<\/p>\n\n\n\n

The answer is yes. <\/p>\n\n\n\n

Springer Nature, for example, has invested<\/a> in research integrity experts and AI-powered tools to help editors and peer reviewers \u201ckeep fraudulent, misleading, or manipulated material out of the research record.\u201d One of these tools, Geppetto, has been developed to detect bogus content generated by AI.\u00a0<\/p>\n\n\n\n

These types of specialised AI detectors are obviously more advanced and sophisticated<\/strong>. This brings us to the risks of making an academic submission based on the false confidence coming from a favorable score from general AI detectors, which are not as sophisticated as the journals’ proprietary tools.<\/p>\n\n\n\n

Even if unintentional and due to oversight, beyond rejected submissions, the author may face accusations of plagiarism or academic misconduct<\/strong> if they represent AI-generated content as their own. This can seriously damage their reputation, academic standing, and career. <\/p>\n\n\n\n

What is the best approach to academic submission? <\/h2>\n\n\n\n

There\u2019s no shortage of AI detectors and AI humanisers, which promise to make AI content undetectable. The internet is also replete with (sometimes unreliable) giveaway signs of AI-generated content, intended to help users modify their work so that it is not flagged as AI-generated. <\/p>\n\n\n\n

Although the benefits of generative AI are undeniable in many areas, there is a simpler approach that frees us from the need to rely on such tools and tips to ensure the originality of our academic work: writing it manually<\/strong>. <\/p>\n\n\n

\n
\"\"<\/figure><\/div>\n\n\n

This approach guarantees the authenticity of your work while strengthening your critical thinking, writing skills, and intellectual integrity. <\/p>\n\n\n\n

While generative AI and AI humanisers offer speed and convenience, they detract from the core of academic writing\u2014originality and intellectual contribution. <\/p>\n\n\n\n

But what if an author lacks confidence in their writing skills<\/strong> and believes that their work does not meet the language standards expected of an academic paper? <\/p>\n\n\n\n

This is where editing assistants that use advanced language technologies, including non-generative AI<\/strong>, come into play. <\/p>\n\n\n

\n
\"\"<\/figure><\/div>\n\n\n

These types of tools do not generate new content, but only improve existing content\u2014helping users maintain their unique voice, originality, and authorship while enhancing language accuracy and quality.<\/p>\n\n\n\n

The scope of their edits falls within \u201cAI-assisted copy editing<\/strong>,\u201d which Springer defines<\/a> as \u201cAI-assisted improvements to human-generated texts for readability and style, and to ensure that the texts are free of errors in grammar, spelling, punctuation and tone.\u201d<\/p>\n\n\n\n

These improvements, as it says, may involve wording and formatting changes, but do not include \u201cgenerative editorial work and autonomous content creation.\u201d Springer does not require authors to declare AI-assisted copy editing. <\/p>\n\n\n

\n
\"\"<\/figure><\/div>\n\n\n

InstaText<\/strong>, an advanced editing assistant<\/strong>, is one of these tools that improves your texts by enhancing clarity, readability, flow, structure, style, tone, conciseness, word choice, grammar, spelling, punctuation, and more\u2014without compromising originality and authorship.<\/p>\n\n\n\n

If you\u2019re interested in submitting apolished and impactful original paper<\/strong> and increasing your chances of getting published<\/strong> while improving your writing and cognitive skills, try InstaText for free<\/a> and see how it can support your academic endeavors. If you\u2019re already a user, be sure to visit our blog<\/a>, which is regularly updated with new information and insights.<\/p>\n\n\n

\n
\"\"<\/figure><\/div>\n\n\n

See also: How to make your writing more readable for academic success<\/a><\/p>\n\n\n\n

\u201cInstaText is a great tool! I use it to improve English texts such as articles, projects and abstracts for conferences. The tool provides very useful suggestions that help me to translate the text to a professional level so that no additional review by \u201cnative speakers\u201d is required. The time and money savings are obvious. I highly recommend it!\u201d

\u2014 Janez Konc<\/strong>, Senior Researcher<\/p>\n\n\n\n

\u201cI used to be able to submit at most one paper a year. Now I can write at least 8 academic papers in less than a year!\u201d

\u2014 Dr. Saeid Fahimeh<\/strong><\/strong>, Researcher<\/p>\n\n\n\n

\u201cInstaText makes your text engaging to read, coherent, and professional-looking. Further, I feel that paragraphs corrected by InstaText look akin to what I see in top marketing and social psychology journals. It is a huge help for an academic writer because rather than focusing on making the text appealing, you can simply focus on what you want to say and build a logically unfolding narration.\u201d

\u2014 Dr. Micha\u0142 Folwarczny<\/strong><\/strong>, Postdoctoral Researcher<\/p>\n","protected":false},"excerpt":{"rendered":"

Learn why relying on AI detectors to ensure the originality of your work can be risky, and how InstaText helps you meet authorship criteria.<\/p>\n","protected":false},"author":3,"featured_media":1761558,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[],"tags":[],"class_list":["post-1761590","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry"],"acf":[],"_links":{"self":[{"href":"https:\/\/instatext.eu\/api\/wp\/v2\/posts\/1761590","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/instatext.eu\/api\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/instatext.eu\/api\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/instatext.eu\/api\/wp\/v2\/users\/3"}],"replies":[{"embeddable":true,"href":"https:\/\/instatext.eu\/api\/wp\/v2\/comments?post=1761590"}],"version-history":[{"count":5,"href":"https:\/\/instatext.eu\/api\/wp\/v2\/posts\/1761590\/revisions"}],"predecessor-version":[{"id":1761620,"href":"https:\/\/instatext.eu\/api\/wp\/v2\/posts\/1761590\/revisions\/1761620"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/instatext.eu\/api\/wp\/v2\/media\/1761558"}],"wp:attachment":[{"href":"https:\/\/instatext.eu\/api\/wp\/v2\/media?parent=1761590"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/instatext.eu\/api\/wp\/v2\/categories?post=1761590"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/instatext.eu\/api\/wp\/v2\/tags?post=1761590"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}