IB Historical Investigation - How to use AI
Is the extended research essay dead?
Picture a university library in 1989. A student sits surrounded by index cards, each one carefully copied by hand from a source he has tracked down over several weeks. Some of those sources took months to locate. One arrived by interlibrary loan from a university from the otherside of the country. He has read everything available on his topic in this library. Access to information was genuinely scarce, and the research essay rewarded the students who could find it, organise it and argue with it.
Is the extended research essay dead?
Picture a university library in 1989. A student sits surrounded by index cards, each one carefully copied by hand from a source he has tracked down over several weeks. Some of those sources took months to locate. One arrived by interlibrary loan from a university from the otherside of the country. He has read everything available on his topic in this library. Access to information was genuinely scarce, and the research essay rewarded the students who could find it, organise it and argue with it.
That 1989 student was me. The essay I was writing was not so different in form from the one you are being asked to write now. And yet the world in which you are researching it could hardly be more different. Each generation since has faced a new technology that their teachers feared would make serious research impossible. The first generation, mine, worked with physical libraries and handwritten notes. There were no computers in my world as far as I could see. I even had to pay someone to type up my thesis. Information was hard to find and that scarcity shaped everything.
The second generation arrived in schools in the late 1990s, when the internet changed the research essay permanently. Mrs Regehr belongs to that generation. For her, finding information was never the problem. The challenge was evaluating it, managing the sheer volume of it, and resisting the temptation to copy and paste rather than think. Schools adapted. Students learned source evaluation, academic honesty and digital literacy. The essay survived.
The third generation is yours. On 30 November 2022, OpenAI released ChatGPT to the public. Within two months it had one hundred million users. Nothing in the history of technology had spread so fast. And unlike the internet, which changed where students found information, AI changed what students could do with it. For the first time, a student could interrogate a historical debate conversationally, summarise competing interpretations in seconds, and produce a plausible-looking essay without engaging seriously with a single source. Will the essay survive? Probably not.
The challenge is not to avoid AI. It is to understand what it can and cannot do, and to use it in ways that strengthen your research rather than replace it. That is what this section of the site is about.
The second generation arrived in schools in the late 1990s, when the internet changed the research essay permanently. Mrs Regehr belongs to that generation. For her, finding information was never the problem. The challenge was evaluating it, managing the sheer volume of it, and resisting the temptation to copy and paste rather than think. Schools adapted. Students learned source evaluation, academic honesty and digital literacy. The essay survived.
The third generation is yours. On 30 November 2022, OpenAI released ChatGPT to the public. Within two months it had one hundred million users. Nothing in the history of technology had spread so fast. And unlike the internet, which changed where students found information, AI changed what students could do with it. For the first time, a student could interrogate a historical debate conversationally, summarise competing interpretations in seconds, and produce a plausible-looking essay without engaging seriously with a single source. Will the essay survive? Probably not.
The challenge is not to avoid AI. It is to understand what it can and cannot do, and to use it in ways that strengthen your research rather than replace it. That is what this section of the site is about.
Using AI
The books and articles you need for your Historical Investigation are almost all available online. JSTOR, Google Scholar, university digital archives and national libraries have between them made available more historical scholarship than any physical library could ever hold. In that sense you are not so different from Mrs Regehr's generation. What AI changes is how you navigate that abundance. You can use it to identify which historians matter for your topic, to understand how a historical debate has developed, to find connections between arguments you might have missed, and to check whether you have understood a difficult piece of academic writing correctly. But it is a starting point, not a finishing line. AI will tell you what historians have argued. It will not tell you who is right. That judgement is yours. It is also, not coincidentally, what the examiner is looking for. Reading slowly and thinking carefully remain the irreplaceable core of good historical research.
But there is a second way AI can help you, and this one is qualitatively new. Previous generations of professional historians were limited not only by what sources existed, but by what they could physically access and process. A researcher in 1989 might spend a year in an archive reading handwritten documents in a foreign language. A researcher in 2004 might find those same documents digitised online but still face the same barriers of language, volume and legibility. Most sources, for most historians, remained effectively out of reach. You face none of those barriers in the same way. AI can now translate primary sources from almost any language in seconds. It can transcribe handwritten documents, process large volumes of material that no individual could read in a lifetime, and help you identify patterns and significance across sources that would previously have required a team of researchers and years of work. This means that sources which were previously ignored, overlooked or simply inaccessible are now open to you. Diplomatic correspondence. Local government records. Personal diaries. Newspapers from countries whose languages you do not speak. For the first time, a secondary school student can do something that genuinely resembles what professional historians do. Not summarise what others have found, but to actually investigate. That is the real opportunity AI offers you.
But there is a second way AI can help you, and this one is qualitatively new. Previous generations of professional historians were limited not only by what sources existed, but by what they could physically access and process. A researcher in 1989 might spend a year in an archive reading handwritten documents in a foreign language. A researcher in 2004 might find those same documents digitised online but still face the same barriers of language, volume and legibility. Most sources, for most historians, remained effectively out of reach. You face none of those barriers in the same way. AI can now translate primary sources from almost any language in seconds. It can transcribe handwritten documents, process large volumes of material that no individual could read in a lifetime, and help you identify patterns and significance across sources that would previously have required a team of researchers and years of work. This means that sources which were previously ignored, overlooked or simply inaccessible are now open to you. Diplomatic correspondence. Local government records. Personal diaries. Newspapers from countries whose languages you do not speak. For the first time, a secondary school student can do something that genuinely resembles what professional historians do. Not summarise what others have found, but to actually investigate. That is the real opportunity AI offers you.
Where to start
The table below lists a small selection of online archives where you can find primary sources. It is not exhaustive. Part of the research process is finding archives relevant to your specific topic. Use it as a starting point.
The table below lists a small selection of online archives where you can find primary sources. It is not exhaustive. Part of the research process is finding archives relevant to your specific topic. Use it as a starting point.
Archive |
What it contains |
Language |
Key political and diplomatic documents, ancient to 21st century |
English |
|
Digitised materials from European cultural institutions, documents, photographs, newspapers |
Multiple |
|
Books, newspapers, films, recordings, out of copyright texts |
Multiple |
|
Digitised Swiss journals and periodicals from the 19th and 20th centuries |
French, German, Italian |
|
American historical documents, photographs, maps, newspapers |
English |
|
French historical documents, newspapers, photographs, manuscripts |
French |
Once you have found a source, AI can help you with the things that would previously have stopped you in your tracks. A document in Welsh, German or Russian. A handwritten letter that is difficult to read. A collection so large you would not know where to begin. Paste the text, upload the image, describe what you are looking for. Ask the AI to translate, transcribe or summarise. Then read the result carefully yourself and decide what it means.
One thing worth keeping in mind as you search. Your Historical Investigation requires you to analyse two sources in depth. They need to serve different purposes and reflect different perspectives. A published secondary source from a respected historian is useful. But a primary source you have found and processed yourself, something in a foreign archive, a contemporary newspaper report, a diplomatic telegram, a handwritten diary entry, tells the examiner something more important. It tells them that you have actually done historical research. It is exactly what the highest marks are looking for.
One thing worth keeping in mind as you search. Your Historical Investigation requires you to analyse two sources in depth. They need to serve different purposes and reflect different perspectives. A published secondary source from a respected historian is useful. But a primary source you have found and processed yourself, something in a foreign archive, a contemporary newspaper report, a diplomatic telegram, a handwritten diary entry, tells the examiner something more important. It tells them that you have actually done historical research. It is exactly what the highest marks are looking for.