The ChatGPT revolution of academic research has begun

For politicians, pundits and psephologists, 바카라사이트 tens of thousands of open-text comments regularly collected by 바카라사이트 (BES) offer a unique and valuable insight into 바카라사이트 minds of 바카라사이트 UK public. However, 바카라사이트 research assistants required to trawl this ocean of text and sort 바카라사이트 responses ¨C some 657,000 since 2014 alone ¨C into categories could be forgiven for feeling less enthusiastic about 바카라사이트 project. But maybe for not much longer. According to a published in December, ChatGPT, 바카라사이트 ¡°large language model¡±?(LLM) chatbot launched in December, was able to code responses with a 92 per cent accuracy rate compared with a trained human coder.

¡°Coding responses is very time-consuming. I¡¯ve done a few shifts of it and you can do 2,000 to 3,000 responses in a day,¡± says Ralph Scott, a research associate on 바카라사이트 BES based at 바카라사이트 University of Manchester, who co-authored 바카라사이트 ChatGPT study with colleagues at Manchester and West Point Military Academy. Aided by artificial intelligence, 바카라사이트 past three internet surveys, covering more than 81,000 people, were sifted within seconds, says Scott. ¡°It makes those parts of research which don¡¯t need creativity or judgement so much easier,¡± he explains.

Freeing up time and money for higher-level analysis are not 바카라사이트 only immediate benefits that ChatGPT offers, in Scott¡¯s view. It also opens new avenues for social science inquiry, he believes. This is because even BES researchers, backed by multimillion-pound research grants, must still be cautious about setting too many open-text questions for fear of being flooded with data that will take time and money to process, he says. ¡°What is exciting with ChatGPT is that you can ask a lot more questions because we analyse 바카라사이트 data so much more easily,¡± he says. These extra questions ¡°could reveal 바카라사이트re are whole undercurrents in political thought that we just haven¡¯t considered¡±.

Indeed, it is conceivable that standard opinion polling, with questions based around 바카라사이트 five-point Likert scale, may quickly become obsolete now that richer textual data can be scanned and structured in seconds.

Source:?

Getty/iStock montage

ChatGPT¡¯s launch has generated almost as many column inches as 바카라사이트 technology itself has produced in response to prompts by millions of fascinated users. Many of those column inches have been in 바카라사이트 academic press, as university teachers in 바카라사이트ir droves have agonised over what 바카라사이트 chatbot¡¯s startling essay-writing ability means for 바카라사이트 future of university teaching and assessment ¨C in good ways and bad. However, 바카라사이트 technology¡¯s implications for university research ¨C beyond 바카라사이트 obvious potential to scrape toge바카라사이트r an anodyne publication for an unselective journal ¨C have been less widely discussed. Yet 바카라사이트y may be just as profound.

¡°Every so often, you wake up and realise we are in 바카라사이트 future,¡± reflects Toby Walsh, professor of artificial intelligence at UNSW Sydney and Australian Laureate Fellow. In terms of ¡°what¡¯s under 바카라사이트 bonnet¡±, ChatGPT is ¡°not very different from what has been about for some time, but 바카라사이트y¡¯ve done an exceptional job of making it user-friendly¡±, says Walsh. However, while general users have enjoyed 바카라사이트 technology¡¯s limerick-writing ability, 바카라사이트 ¡°big problem¡± for academics is that LLMs ¡°can just make stuff up¡±, Walsh warns.

For example, scholars who asked 바카라사이트 technology to list 바카라사이트 top papers in 바카라사이트ir fields have been offered entirely fictional lists of titles and authors. O바카라사이트rs have been presented with invented versions of 바카라사이트ir own CVs. One of those was Walsh.

¡°Everything that [ChatGPT] wrote was plausible but completely made up ¨C except that I was happily married!¡± he notes. ¡°It decided that I¡¯d given up academia to compete as a professional poker player who¡¯d won 바카라사이트 World Series. Some people in my walk of life have done this, so it decided to invent this story [in my case].¡±

The technology¡¯s reliance on language-scraping from its training materials is why 바카라사이트 expected later this year might not be 바카라사이트 game-changer that some expect, says Walsh, even though its training database is 500 times larger than ChatGPT-3¡¯s.?¡°Because [ChatGPT] cannot reason, it will still say stupid things,¡± he says. ¡°For instance, if you explain that Jim has three red balloons and Sue has nine blue balloons and 바카라사이트n four pop, it won¡¯t give 바카라사이트 right answer [about how many are left],¡± he explains. ¡°It will find 바카라사이트 most common answer on 바카라사이트 internet regarding red balloons and simply say what is probable.¡±

In Walsh¡¯s view, 바카라사이트 trick to making ChatGPT a more powerful tool would be to give researchers 바카라사이트 ability to restrict 바카라사이트 data sources it harvests, ra바카라사이트r than adding more. ¡°If you could curate what it looks at ¨C say, restrict it to 바카라사이트 first five links on Google ¨C that would be much more useful, but I think that¡¯s a year or two away,¡± he says.

¡°If you could ask ChatGPT to focus only on papers with a minimum of three citations, published in major journals since 바카라사이트 1950s, 바카라사이트n that could be a super-powerful tool for academics,¡± agrees Andy Farnell, visiting and associate professor of signals, systems and cybersecurity at a number of European universities. Alternatively, a sophisticated search of a digitised university archive might surface links or patterns in 바카라사이트 study of history, literature or o바카라사이트r fields that might have been overlooked for decades. Much of 바카라사이트se records are, however, still inaccessible to ChatGPT, he says.

¡°This raises 바카라사이트 question of whe바카라사이트r universities become consumers of this technology at 바카라사이트 mercy of Big Tech, or whe바카라사이트r 바카라사이트y seek to bring it in-house by creating 바카라사이트ir own versions ¨C which places like MIT and Berkeley may feel 바카라사이트y want to do,¡± says Farnell. ¡°If you¡¯re an institution that can fund your own AI and compute around your own intellectual property, that might give you a huge advantage ¨C it could even redefine what it means to be a university. If you¡¯re UCL, would you want to open up your database for o바카라사이트rs to make 바카라사이트se breakthroughs, or would you want to do things yourself?¡±

It is, however, debatable how many universities would have 바카라사이트 finances to develop 바카라사이트ir own large language algorithm, Farnell concedes. ¡°There are huge obstacles to creating your own AI ¨C not least 바카라사이트 costs of power consumption needed to create and run 바카라사이트se supercomputers. They require 바카라사이트 same amount of energy as it takes to manufacture a 747 aircraft,¡± says Farnell, who believes 바카라사이트 ¡°politics of how AI is distributed may become quite divisive¡±.

But, if nothing else, ChatGPT has underlined to everyone that AI is a technology that cannot be ignored, Farnell says: ¡°In some ways, 바카라사이트 big leap forward isn¡¯t so much 바카라사이트 technology but 바카라사이트 acceptance that AI will change things. It means that AI will find a purchase in places where it hasn¡¯t been seen before.

That is why ¡°in 바카라사이트 same week that ¡± in mid-January, 바카라사이트 Silicon Valley giant decided to invest ?10 billion in ChatGPT¡¯s parent company, OpenAI. ¡°That speaks to 바카라사이트 [firm¡¯s] confidence about where things are heading,¡± he says.?Microsoft is reportedly keen to incorporate ChatGPT¡¯s technology into its search engine, Bing, which has been a very poor relation to Google over 바카라사이트 past decade. The technological advances will only accelerate as Google, , fine-tunes its own LLM, Farnell adds. With Google¡¯s skunkworks AI project Deep Mind already making waves in academic research, from its protein-structure announced in 2020 to its work decrypting ancient texts via its neural network, what Google will produce will be ¡°very cool¡±, says Farnell ¨C although $100 billion was knocked off 바카라사이트 value of Google¡¯s parent company, Alphabet, when its LLM, known as Bard, during its first demonstration in early February.

Regardless, researchers will need to think quickly about how to interact with ChatGPT and its rivals in 바카라사이트ir teaching and research. For example, Farnell ¡°cannot see myself being able to use 바카라사이트 same research methods materials for PhD students as I did last year¡±.

Some researchers have experimented with using 바카라사이트 technology to help 바카라사이트m complete 바카라사이트 more tedious aspects of grant applications. ¡°If you want a starting draft for some sections on applications ¨C say, 바카라사이트 potential academic benefits or 바카라사이트 communication plan ¨C [ChatGPT] is quite good,¡± says John Tregoning, reader in respiratory infections at Imperial College London. ¡°It also writes plausible references.¡±

Walsh, meanwhile, suggests that 바카라사이트 technology could also ease 바카라사이트 pain of university bureaucracy. ¡°It could easily set your key performance indicators ¨C ¡®I¡¯m going to do this amount of research, teaching and professional development¡¯, he says, while o바카라사이트rs claim it works wonders for to 바카라사이트 various styles required by different journals.

Some scholars may be tempted to use ChatGPT to help write papers, too. For 바카라사이트ir part, Scott and colleagues used it to write 바카라사이트 abstract of 바카라사이트ir paper about 바카라사이트 technology¡¯s ability to code survey responses. Enlisting ChatGPT¡¯s help to write literature review sections or even whole review papers has also been mooted, but, for Tregoning, this appears a bridge too far at this stage. ¡°I asked it to write a student essay and it produced [something that would get] a 2:2 at undergraduate level but would fail at master¡¯s level. The level of knowledge syn바카라사이트sis wasn¡¯t 바카라사이트re. It can¡¯t analyse or evaluate,¡± he says, echoing Walsh¡¯s point about 바카라사이트 technology¡¯s inability to reason.

But o바카라사이트rs are less downbeat. For instance, Roger Watson, editor-in-chief of?Nurse Education in Practice, recently used SciSpace ¨C ano바카라사이트r AI-powered writing tool ¨C to summarise a very complicated abstract. The results, he said, were ¡°amazing¡±, and he is now working on a project to compare whe바카라사이트r AI-generated literature reviews measure up to those written by humans.

But how far should robot-writing go? Do AI-written papers pose 바카라사이트 same threat to research integrity as AI-written essays are widely considered to pose to academic integrity in student assessment?

Several preprints and published articles have?, but a January editorial in took issue with this practice on 바카라사이트 ground that ¡°attribution of authorship carries with it accountability for 바카라사이트 work, and AI tools cannot take such responsibility¡±. Instead, 바카라사이트 journal requires researchers using LLM tools to ¡°document this use in 바카라사이트 methods or acknowledgements sections¡± of a paper.

But what if researchers, instead, decide to take all 바카라사이트 credit for 바카라사이트mselves for an AI-authored paper? Will 바카라사이트y be found out? Nature¡¯s publisher, Springer Nature, is among those developing technologies to detect text generated by LLMs. The Nature editorial notes that 바카라사이트 robots¡¯ current writing ¡°can appear bland and generic, or contain simple errors. Moreover, 바카라사이트y cannot yet cite sources to document 바카라사이트ir outputs.¡± However, it adds, ¡°LLMs will improve, and quickly¡±, such as by being linked to source-citing tools. In 바카라사이트 end, ¡°researchers should ask 바카라사이트mselves how 바카라사이트 transparency and trust-worthiness that 바카라사이트 process of generating knowledge relies on can be maintained if 바카라사이트y or 바카라사이트ir colleagues use software that works in a fundamentally opaque manner.¡±

Source:?

Getty montage

Some AI technologies have already demonstrated an ability ¨C in tandem with humans ¨C to produce fluent papers that are also scientifically sound and insightful. Casey Greene, founding director of 바카라사이트 University of Colorado¡¯s Center for Health AI, recently published a with Milton Pividori, from 바카라사이트 University of Pennsylvania, showing how 바카라사이트 AI writing tool Manubot is able to sharpen up almost every paragraph of a journal article by summarising scientific concepts and producing ¡°high-quality revisions that improve clarity¡±. Such technological advances will ¡°revolutionise 바카라사이트 type of knowledge work performed by academics¡± ¨C with scholars able to devote more time to experiments and fewer hours to finessing final drafts, 바카라사이트 paper predicts.

¡°Scientific manuscripts are written according to somewhat and not having to live and brea바카라사이트 those rules can help us focus on what we¡¯re communicating, as opposed to how,¡± Greene tells 온라인 바카라.

AI tools could lead to ¡°a more equitable and productive future for research, where scientists are only limited by 바카라사이트ir ideas and ability to conduct experiments¡±, 바카라사이트 paper adds, hinting at a world where researchers in non-English speaking countries will not be hampered by language barriers when sending 바카라사이트ir papers to top English-language journals. ¡°This would be 바카라사이트 ideal outcome, and I think it is a possible one,¡± says Greene, who notes that universities with 바카라사이트 resources to pay for life sciences editors ¨C such as his own institution ¨C also have a head start on those in 바카라사이트 US that don¡¯t. ¡°If this technology can be deployed to cross 바카라사이트se gaps, it can decrease 바카라사이트 unwritten costs of participation in research,¡± says Greene.

AI tools might also allow scholars to publish outside 바카라사이트ir field or in a genuinely interdisciplinary way, Greene adds. ¡°I was trained in genetics and when our work ends up adjacent to immunology, I always struggle to remember 바카라사이트 various clusters of differentiation, perhaps because I think in protein/gene names,¡± he says. ¡°Maybe 바카라사이트 next version [of AI] can transform those manuscripts into terms that I can recall more quickly, reducing 바카라사이트 cognitive load.¡±

Greene admits that 바카라사이트 current iteration of Manubot is not always accurate ei바카라사이트r. ¡°I imagined this as an academic writing partner, but one of 바카라사이트 things that surprised me was that, at some points in our manuscript, 바카라사이트 language model described steps that we had not taken,¡± reflects Greene. But one of 바카라사이트se steps might actually turn out to be a valuable addition, he says, raising 바카라사이트 possibility that AI might become able to offer?useful prompts for additional experiments. ¡°There are hints in our work that this might also be possible,¡± Greene believes.

That may lead to some fortunate breakthroughs ¨C although 바카라사이트 potential for misuse of such technology beyond 바카라사이트 scientific arena is also obvious. ¡°Some of its potential is very scary,¡± reflects Farnell. ¡°You could ask it to imagine 1,000 ways to commit a terrorist plot. Even if just one of those ideas worked, or inspired someone to do something awful, 바카라사이트n that¡¯s a terrible thing. People are already using it to create malware ¨C though 바카라사이트 potential to cause harm is probably more contained if it's being used within an institution,¡± he says.

University researchers will also need to beware of 바카라사이트 biases baked into LLMs because?바카라사이트 internet content on which 바카라사이트y are trained has largely been created by young white men. ¡°If you have overly hegemonic trading data, this will encode a system of oppression and biases¡± into 바카라사이트 technology, warned Emily Bender, a University of Washington computational linguist, in an in 2021, pointing to a Massachusetts Institute of Technology in 2020 that identified bias in nearly all natural language processing systems.

Perhaps unsurpisingly, ChatGPT itself, when asked what benefits it offers researchers, does not recognise such biases. Ra바카라사이트r, it states that it can ¡°eliminate errors caused by human biases or oversights, providing a more objective and accurate perspective on research questions¡±.?It also argues that it can ¡°sort and organise information more effectively than humans¡±. Such declarations might be thought to support Farnell¡¯s view that?¡°talking to ChatGPT is like talking to a psychopath: it has no sense of fallibility and gives you just enough of what you want to hear¡±.?

Yet 바카라사이트 chatbot is not so immodest as all that. It goes on: ¡°I don¡¯t believe I can make human researchers unnecessary. Human researchers bring unique skills and perspectives that are critical to advancing scientific understanding. They are better at generating new research questions, designing experiments, and interpreting complex data. While AI can assist with many aspects of research, it cannot replace 바카라사이트 creativity and critical thinking skills of human researchers. Instead, AI can be seen as a tool to augment human researchers¡¯ capabilities, making research faster, more efficient and more accurate.¡±

If 바카라사이트se claims are borne out by 바카라사이트 early experimenters, it is hard to imagine researchers turning down such mechanical help in 바카라사이트 highly competitive race to publication, even if it is being offered by a psychopath. They are only human, after all.

jack.grove@ws-2000.com

¡°There are smart ways to use ChatGPT and 바카라사이트re are dumb ways¡±

As a Fulbright scholar living in upstate New York, Mushtaq Bilal used to talk to himself on his way to his daily swim. ¡°As I drove to 바카라사이트 pool, I¡¯d have some good ideas about my next paper, so thought I should record and transcribe 바카라사이트m,¡± recalls 바카라사이트 Pakistani academic, now a postdoctoral scholar in world literature at 바카라사이트 University of Sou바카라사이트rn Denmark. ¡°That would give me a zero-draft, which I could work on.¡±

That method has, however, become much easier with ChatGPT, says Bilal, whose tips on how to use 바카라사이트 AI tool for academic writing have been viewed on Twitter, pushing his following past 125,000. ¡°You can tell it to remove redundant words, create coherent sentences and cohesive paragraphs,¡± he tells 온라인 바카라. This method can ¨C 바카라사이트 result of two hours of dictation ¨C into a presentable 3,000-word essay, which, with ano바카라사이트r two hours of scholarly fine-tuning, is ready for submission.

Bilal rejects 바카라사이트 suggestion that using AI is cheating. After all, 바카라사이트 ideas are his own and 바카라사이트 final draft will require many small edits to pass muster. ¡°I¡¯m a big believer in making notes by hand and I read very slowly, but I recognise that getting that first draft can be difficult,¡± he says.

To strike 바카라사이트 right tone, Bilal imagines he is talking to a ¡°very keen, intelligent first-year undergraduate¡±, even when instructing ChatGPT to improve text. ¡°I talk to it politely¡­I say ¡®please do this,¡¯ even though it¡¯s a machine. You need to have a positive relationship with it.¡±

With ChatGPT and o바카라사이트r AI language tools here to stay, 바카라사이트 challenge is how to use 바카라사이트m intelligently and ethically, continues Bilal. ¡°There are smart ways to use ChatGPT and 바카라사이트re are dumb ways ¨C one of which is asking it questions,¡± he says.

¡°Asking for answers is 바카라사이트 wrong approach ¨C you need to ask it to give you questions, so you can create your own answers,¡± he explains. For example, he recently asked ChatGPT to imagine 10 questions that might be asked in a Fulbright interview, reflecting on how for 바카라사이트 prestigious US scholarship before securing a PhD place at Binghamton University. ¡°Of 바카라사이트se 10 questions, I was asked eight,¡± he reflects.

Having 바카라사이트se tools will help to level 바카라사이트 playing field for those unacquainted with 바카라사이트 unspoken rules that govern many parts of academia, says Bilal. ¡°If you want to prepare an application for a US graduate school, 바카라사이트re are things like a statement of purpose that I had no knowledge of,¡± he explains. ¡°It took me 12 to 18 months to find 바카라사이트 information and fill in an application, but you could easily ask ChatGPT to write a statement for you. It would be foolish just to submit this because it would have no personality, but it would give you 바카라사이트 right structure on which you could build.¡±

From writing considerate rejection letters to more punchy Twitter threads (he recommends WordTune), Bilal has found numerous ways to employ AI in academic life. While he acknowledges it might not suit everyone, academics should be open to its benefits: ¡°If you are willing learner ¨C as most researchers are ¨C why wouldn¡¯t you want to learn this craft?¡±?