The current AI landscape
Researchers and clinicians are on a journey from awareness to usage to benefit when it comes to AI.
Attitudes toward AI: Chapter 1
Dive into the evolving AI landscape, where high awareness among researchers and clinicians is gradually translating into practical usage and significant benefits. Uncover detailed insights on familiarity, perceptions and the transformative impact of AI across various regions and professions.
The current AI landscape
Awareness of AI in general is high among both researchers and clinicians, but relatively few say they are currently very familiar with the technology, having used AI a lot.
96% have heard of AI (including GenAI) – subsequent statistics exclude the 4% not familiar with AI
54% have used AI (including GenAI) and 31% have used it for work purposes; this is higher in China (39%) than in the USA (30%) and India (22%)
11% are very familiar with AI (including GenAI), i.e. they’ve used it a lot
ChatGPT is by far the most well-known AI product (89%)
25% have used ChatGPT for work purposes
49% of those who have not used AI cite a lack of time as the reason
72% believe AI (including GenAI) will have a transformative or significant impact on their area of work
Over half of both groups who are aware of AI have used it, and almost one-third have used it for a specific work-related purpose; this is highest in China (39%). A lack of time to investigate such tools is the main reason for not using AI.
Fig 1. Question: To what extent are you familiar with AI (including GenAI)? n=3,118; Question: Have you used an AI (including GenAI) product or an AI feature on a product you use regularly? n=2,999; Question: Do you expect you will choose to use AI (including GenAI) in the near future? n=1,330.
"AI is here, and it works. We need to think how to use it properly."
SRRK
Survey respondent, researcher, Kazakhstan
Awareness of GenAI tools
Given the rapid rise of GenAI tools, particularly ChatGPT, it is perhaps unsurprising that most people are at least aware of its existence. Global Counsel surveys show that about 90% of people in the UK, US and Germany have heard of GenAI.17 Similarly, a 2023 survey by Pew Research Center showed that 90% of Americans say they “have heard at least a little about artificial intelligence.”18 However, drilling down reveals only about one-third have heard a lot about it.
The current survey reflects these findings: almost all (96%) have heard of AI. Awareness is highest in China at 99% (see accompanying databook for full details). Globally, only 11% are very familiar with AI, having used it a lot, this is higher among researchers (14%) than it is clinicians (8%). Demographics seem to have an impact when it comes to familiarity with AI. We see that in APAC those very familiar and using AI a lot is highest 13%, compared to 8% in Europe (see accompanying databook for full details). Please note: all subsequent statistics in this report exclude the 4% not familiar with AI.
ChatGPT is the most familiar GenAI tool
ChatGPT is by far the most well-known AI product, with 89% of survey respondents globally being familiar with it. Researchers (94%) are more likely than clinicians (84%) to have heard of it.
Fig 2. Question: To what extent are you familiar with AI (including GenAI)?
Fig 3. Question: Which of these AI products, if any, have you heard of before today? n=2,999; Question: Which, if any, AI products or AI features have you used for work purposes? Shown to only those aware of tool base variable n=81 to 2,713: % shown is proportion of 2,999.
This reflects the rapidly increasing awareness and use of ChatGPT. In March 2023, just under a year before the current survey, Pew Research found that 58% of US adults had heard of ChatGPT.19 In a consumer survey by Capgemini Research Institute the same year, 51% of respondents were aware of the latest trends in GenAI and had explored tools like ChatGPT and DALL-E.20 A further 35% were aware but had not yet tried the tools.
In the current survey, we found that across most geographical regions, ChatGPT was the most well-known tool. Familiarity was high across all regions but notably lower in South America at 75% (see accompanying databook for full details).
The next most familiar GenAI tool is Bard (40% overall), followed by Bing Chat (39%), Gemini (22%) and MS Copilot (22%). Lesser-known tools are Semantic Scholar (17%), ChatPDF.ai (13%) and OpenEvidence (8%). In most cases, researchers are more likely than clinicians to be familiar with the tools.
The most well-known GenAI tools we asked participants about
ChatGPT – a chatbot developed by OpenAI
Gemini – formerly known as Bard, a chatbot developed by Google
Copilot – formerly Bing Chat, an AI-powered feature by Microsoft, built into the browser, and Microsoft 365 Copilot, built into Microsoft 365 apps including Word, Excel, PowerPoint, Outlook and Teams
Semantic Scholar – an AI-powered research tool for scientific literature, developed at the Allen Institute for AI
ChatPDF.ai – AI-powered text recognition, table extraction and data analysis for PDFs
OpenEvidence – an AI system to aggregate, synthesize and visualize clinically relevant evidence
Perceptions of AI
While awareness of certain GenAI tools is high among both researchers and clinicians, attitudes to the technology are more variable, with 49% of respondents globally saying they feel mixed about AI, able to see both potential and drawbacks.
However, sentiment is generally more positive about the impact of AI than negative: 36% of respondents say AI is a welcome advancement, compared to just 1% who see mostly drawbacks. Researchers (41%) are more positive about the technology than clinicians are (32%). Clinicians are also more unsure, with 17% saying they need to see how AI develops, compared to 10% of researchers.
A report by Stanford University highlighted major differences in sentiment about AI, with that of people in Asia more positive than those in the West.21 In this survey, we see that those in North America are more skeptical, 28% are positive about AI, this compares to 40% in APAC and 46% in China (see accompanying databook for full details). A similar pattern is evident in Pew Research Center Survey in the US, 52% of adults reported being more concerned than excited about the use of AI in daily life.18 Previous research conducted by the UK’s Office for National Statistics also reveals a mixed picture when it comes to perceptions.22 In a 2023 survey of adults, 28% said they think “AI brings greater risks than benefits,” compared to just 14% thinking the opposite.
Men have also been shown to be more positive than women about AI, and this is corroborated in the current survey; men are much likely to be positive about the impact of AI (45%) than women (27%) (see detailed findings in databook).
There is little evidence of differences in sentiment by age group, however, there may be differences by sector and role. In a survey by Gapgemini, executives have a more positive outlook: 74% “believe the benefits that generative AI brings outweigh the associated risks,” rising to 80% of executives in pharma and healthcare companies.23
“All emerging technologies, including AI, have both advantages and disadvantages. It is essential to further develop and regulate these technologies, aiming to extract maximum benefits.”
SRRC
Survey respondent, researcher, Canada
Fig 4. Question: What are your overall feelings about the impact of AI on your area of work?
AI will have a major impact
The vast majority (95%) of respondents who are aware of AI believe it will have an impact on their work (1% think it won’t, 4% don’t know), with 72% believing the level of impact will be either transformative or significant. There is some variation in the expected extent of that impact, with a higher proportion of researchers (28%) than clinicians (22%) expecting the impact to be transformative.
These results are echoed in other research. In View from the Top: Academic Leaders’ and Funders’ Insights on the Challenges Ahead, AI was identified as “a transformative force that will affect all aspects of university functions, from teaching and research to administration,” with some respondents expecting AI to have a “profound impact” across disciplines.24
Capgemini research also reveals the sentiment that GenAI could transform work.23 In their survey, 70% of consumers said they believe GenAI will make them more efficient at work and free up time to be more strategic, while 70% of executives agree GenAI will allow organizations to widen knowledge workers’ roles. And 60% say GenAI will completely revolutionize the way they work.
Like much of the other research, the Network Readiness Index (NRI), which looks at countries’ digital readiness, highlighted a mixed perception of GenAI among the public.25 The NRI shows a geographical split, which shows the USA ranks highest in terms of digital readiness, largely due to their pioneering position in technology, followed by Switzerland and Hong Kong.25 However, it’s the Republic of Korea, Israel and Japan that lead the ‘people’ pillar of digital readiness, which comprises individuals, businesses and governments.
The current study somewhat reflects this distribution with more in APAC believing the impact of AI will be transformative or significant (76%) versus North America (66%) and Europe (65%) (see accompanying databook for full details). This regional pattern was also evident in Clinician of the Future 2023, in which clinicians in China were least likely to find the future use of AI undesirable (17%), compared to 33% in Europe and 31% in North America.26
Fig 5. Question: What do you think will be the level of impact of AI (including GenAI) in your area of work in the near future?
Men and women differ in AI views
Men are significantly more likely than women to feel positive about AI, at 45% and 27% respectively. Women are more likely than men to feel mixed (54% versus 44%) and unsure (17% versus 10%).
This difference is reflected in their expectations: men are more likely than women to think AI will be transformative, at 29% and 21% respectively. Overall, more women (26%) than men (20%) expect the change resulting from AI to be only partial.
Conversely, women are more concerned about the ethical implications of AI on their area of work, with 30% of women reporting significant concerns, compared to 24% of men.
Women also appear to be more aware of their institutions’ actions related to AI, with women less likely than men to be unsure how their institution is preparing for AI usage, and more women than men likely to be aware of new AI leadership at their institutions.
AI in practice
Of those familiar with AI, more than half (54%) of respondents in the current research have actively used AI, with researchers (59%) more likely than clinicians (50%) to have used it.
Proportionally more researchers (37%) than clinicians (26%) have used AI tools for a work-related purpose.
While this survey is setting a baseline, there are indications of an upward trend. In the Research Futures 2.0 report published in 2022, 8% of respondents reported using AI extensively in their research.27 The researchers who used AI reported using it to analyze research results (66%), process large data sets to spot defects (49%), help conduct research (36%), enhance images (26%) and generate hypotheses (17%).
Elsevier’s Clinician of the Future 2023 report noted that AI technology is already helping clinicians learn and make decisions – not displacing or replacing them, but supplementing and supporting them.3 Clinicians surveyed were open to the potential of AI to improve patient care, though in practice this was limited at the time: respondents shared that 11% of their clinical decisions were assisted by GenAI, with nurses (16% of clinical decision) using it more than doctors (7% of clinical decisions).
When we look at the results at a more granular level, regional differences emerge, APAC is more likely to have used AI for work related purposes (34%) versus North America (30%), and it is notably higher in China (39%). There are difference by years of experience in work and gender too. Those active 6-10 years have used AI for a work-related purpose most, while those most experienced are less likely to have used AI than average, while men (35%) are more likely than women 27% to have used AI for work purposes (see detailed findings in databook).
Elsevier’s Research Futures 2.0 report in 2022 revealed similar differences in the extensive use of AI in research.
Fig 6. Question: Have you used an AI (including generative AI) product or an AI feature on a product you use regularly?
Years’ experience and perceptions of AI
Years of experience in work have an impact on people’s perception of AI and knowledge of their institutions’ AI-related actions. As years active in their area of work increases, so too does the proportion who feel unsure about AI. Those newest to their roles tend to be more positive about AI, and they believe it will free up time for higher value work and increase work consistency (see detailed findings in databook). Those with over 35 years’ experience in their area of work are least likely to think AI will be transformative, at 19%, and most likely to expect only a partial change, at 28%. This group is also more likely to be unsure how their institution is preparing for AI usage.
In terms of tools being used, it is not surprising given its high awareness levels that ChatGPT is the AI tool used most for work (see figure 3). Specifically, one-quarter (25%) of respondents in the current study have used ChatGPT for work, with usage significantly higher among researchers than clinicians (31% vs 19% respectively). Comparatively few (4%) report using MS Copilot (in Word, Excel and PowerPoint).
Fig 7. Question: Which of the following describes why you haven’t used an AI product or AI feature?
Why not use AI?
The most common reason for researchers and clinicians not having used AI is lacking the time to investigate or experiment with the tools – 49% of respondents globally cite this as the reason, including 52% of researchers and 47% of clinicians.
Other reasons given for not having used AI tools include lack of access (26%), not having the right tools (25%) and having concerns about AI tools (22%).
More than one in ten (12%) overall say they haven’t used AI due to restrictions, for example, from their employer, funder or publisher.
There are some differences by region: we see that more respondents in South America have not used AI tools due to a lack of subscriptions to such tools (29% vs 20% in North America) and restrictions are highest in North America (15%) which is higher than Europe (9%) (see detailed findings in databook).
Although only 2% of respondents are prohibited from using AI in any way (see figure 8), 27% report being prohibited from uploading confidential information to public GenAI platforms, 18% are prohibited from using it for certain purposes and 10% are prohibited from using certain tools. More than one-quarter (26%) say they are prohibited from using AI tools due to lack of budget to pay for them. A lack of budget is the most commonly cited restriction, at 35% in South America and Middle East and Africa. One-third (33%) of respondents are unaware of restrictions on AI usage at their institution.
Many respondents are also unaware of their institutions’ plans when it comes to AI, with 44% of respondents not knowing how their institutions are preparing for AI usage (see chapter 2, page 29).
Fig 8. Question: Which restrictions, if any, does your institution currently have with regards to AI usage?
Learn more about Attitudes toward AI
References
17. Portulans Institute. Network Readiness Index 2023. Page 19. https://download.networkreadinessindex.org/reports/nri_2023.pdf 새 탭/창에서 열기
18. Michelle Faverio and Alec Tyson. What the data says about Americans’ views of artificial intelligence. Pew Research Center. 21 November 2023. https://www.pewresearch.org/short-reads/2023/11/21/what-the-data-says-about-americans-views-of-artificial-intelligence/ 새 탭/창에서 열기
19. Emily A. Vogels. A majority of Americans have heard of ChatGPT, but few have tried it themselves. Pew Research Center. 24 May 2023. https://www.pewresearch.org/short-reads/2023/05/24/a-majority-of-americans-have-heard-of-chatgpt-but-few-have-tried-it-themselves/ 새 탭/창에서 열기
20. Capgemini Research Institute. Why Consumers Love Generative AI. 7 June 2023. https://prod.ucwe.capgemini.com/wp-content/uploads/2023/06/GENERATIVE-AI_Final_WEB_060723.pdf 새 탭/창에서 열기
21. Stanford University. SQ6. How has public sentiment towards AI evolved, and how should we inform/educate the public? One Hundred Year Study on Artificial Intelligence (AI100). 2021. https://ai100.stanford.edu/gathering-strength-gathering-storms-one-hundred-year-study-artificial-intelligence-ai100-2021-1/sq6 새 탭/창에서 열기
22. Office for National Statistics. Public awareness, opinions and expectations about artificial intelligence: July to October 2023. https://www.ons.gov.uk/businessindustryandtrade/itandinternetindustry/articles/publicawarenessopinionsandexpectationsaboutartificialintelligence/julytooctober2023 새 탭/창에서 열기
23. Capgemini Research Institute. Harnessing the Value of Generative AI: Top Use Cases Across Industries. July 2023. https://prod.ucwe.capgemini.com/wp-content/uploads/2023/07/GENERATIVE-AI_-Final-Web-1-1.pdf 새 탭/창에서 열기
24. Elsevier. View from the Top: Academic Leaders’ and Funders’ Insights on the Challenges Ahead. March 2024.
25. Portulans Institute. Network Readiness Index 2023. https://download.networkreadinessindex.org/reports/nri_2023.pdf 새 탭/창에서 열기
26. Elsevier. Clinician of the Future 2023. Page 27.
27. Elsevier. Research Futures 2.0. 새 탭/창에서 열기 April 2022. Page 64.