On Chatbots and Generative Artificial Intelligence

Oermann, Eric K.; Kondziolka, Douglas

doi:10.1227/neu.0000000000002415

Cited by 22 publications

(17 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Similar trends can be noted for share of skilled workers as well. Another readiness study conducted by UNCTAD [39] also suggests that the developing countries generally encounter problems while adapting, adopting, and using frontier technologies such as research and development, digital infrastructure, and skills, in our case LLMs such as chatGPT. Aforementioned was the case from technological and development perspective, however, digital divide is also created amongst students due to the available Internet speed.…”

Section: Digital Dividementioning

confidence: 82%

“…Since the launch of chatGPT, it has been clear that the platform can boost the creativity and productivity of students, teachers, researchers, content creators, and others. From the perspective of development, it is yet to be observed that who will benefit the most from chatGPT, and how it will impact the low-income countries and workers in the Global South [39]. However, there is no denying that chatGPT is more affordable in comparison to human-like AI assistants such as Google Assistant, Alexa, and Siri as they require google devices, echo dot, and iPhone, whereas chatGPT requires internet access and basic literacy level.…”

Section: Digital Dividementioning

confidence: 99%

“…It is also predicted that most workers and firms operating in developing countries will not be able to make the most from LLMs such as chatGPT due to the unavailability of high-speed internet and high-skilled labour, thus, creating a digital divide between high-income and low-income countries. A study from United Nations Conference on Trade and Development (UNCTAD) [39], provided the data regarding number of skilled workers and mean download speed (Mbps) and suggested that lowincome or upper-low income countries lag in the high-speed internet as well as share of skilled workers, therefore, they are slower in adoption of digital technologies. A visual illustration of their study is shown in Figure 4.…”

Section: Digital Dividementioning

confidence: 99%

See 2 more Smart Citations

Digital Information Literacy Skills among Library and Information Science Professionals in University Libraries of Sindh Pakistan

Ali¹

2022

JIMP

View full text Add to dashboard Cite

This study aims to prevail the digital information literacy skills among library and information professionals working in different universities of Sindh, Pakistan. For data collection, survey method, and a questionnaire comprised of close-ended questions were used. The population of the study consisted of 138 library professionals working in different public and private universities and institutes of Sindh, a province of Pakistan. The results of study show that the greater part of respondents had good digital information literacy skills regarding the use, understanding, evaluation and communication the digital information. With this, majority of the participants were either self-taught or learn through conferences, workshops, friends and colleagues. The barriers in obtaining digital information literacy skills were found; lack of funds, un-updated LIS curriculum, tough working routine, lack of support from higher authorities. Moreover, this study provides an understanding as to how universities and institutes can strength and up-skill their library professionals.

show abstract

Section: Digital Dividementioning

confidence: 82%

Section: Digital Dividementioning

confidence: 99%

Section: Digital Dividementioning

confidence: 99%

See 1 more Smart Citation

Digital Information Literacy Skills among Library and Information Science Professionals in University Libraries of Sindh Pakistan

Ali¹

2022

JIMP

View full text Add to dashboard Cite

show abstract

“…It is also predicted that most workers and firms operating in developing 6 https://unctad.org/meeting/cstd-side-event-public-lecture-professorrichard-baldwin countries will not be able to make the most from LLMs such as chatGPT due to the unavailability of high-speed internet and high-skilled labour, thus, creating a digital divide between high-income and low-income countries. A study from United Nations Conference on Trade and Development (UNCTAD) [39], provided the data regarding number of skilled workers and mean download speed (Mbps) and suggested that lowincome or upper-low income countries lag in the high-speed internet as well as share of skilled workers, therefore, they are slower in adoption of digital technologies. A visual illustration of their study is shown in Figure 5.…”

Section: Digital Dividementioning

confidence: 99%

ChatGPT Needs SPADE (Sustainability, PrivAcy, Digital divide, and Ethics) Evaluation: A Review

Khowaja¹,

Khuwaja²,

Dev³

2023

Preprint

View full text Add to dashboard Cite

<p>ChatGPT is another large language model (LLM) inline but due to its performance and ability to converse effectively, it has gained a huge popularity amongst research as well as industrial community. Recently, many studies have been published to show the effectiveness, efficiency, integration, and sentiments of chatGPT and other LLMs. In contrast, this study focuses on the important aspects that are mostly overlooked, i.e. sustainability, privacy, digital divide, and ethics and suggests that not only chatGPT but every subsequent entry in the category of conversational bots should undergo Sustainability, PrivAcy, Digital divide, and Ethics (SPADE) evaluation. This paper discusses in detail about the issues and concerns raised over chatGPT in line with aforementioned characteristics. We support our hypothesis by some preliminary data collection and visualizations along with hypothesized facts. We also suggest mitigations and recommendations for each of the concerns. Furthermore, we also suggest some policies and recommendations for AI policy act, if designed by the governments. </p>

show abstract

“…Artificial intelligence (AI) systems promise many potential applications in medicine, such as differential diagnosis generation and selection, clinical decision support, and analysis of imaging-, physiologic-, and genomic-based data. 1,2 Within this discipline, attention has grown around ChatGPT (OpenAI; San Francisco, CA), a general Large Language Model developed by OpenAI and initially launched for public use in November 2022. ChatGPT, also known as GPT-3.5, was trained on a large corpus of text data through a combination of supervised and unsupervised learning techniques, followed by fine-tuning via reinforcement learning with human feedback.…”

Section: Introductionmentioning

confidence: 99%

Performance of ChatGPT and GPT-4 on Neurosurgery Written Board Examinations

Ali

Tang

Connolly

et al. 2023

Preprint

View full text Add to dashboard Cite

Background: Interest surrounding generative large language models (LLMs) has rapidly grown. While ChatGPT (GPT-3.5), a general LLM, has shown near-passing performance on medical student board examinations, the performance of ChatGPT or its successor GPT-4 on specialized exams and the factors affecting accuracy remain unclear. Objective: To assess the performance of ChatGPT and GPT-4 on a 500-question mock neurosurgical written boards examination. Methods: The Self-Assessment Neurosurgery Exams (SANS) American Board of Neurological Surgery (ABNS) Self-Assessment Exam 1 was used to evaluate ChatGPT and GPT-4. Questions were in single, best- answer, multiple-choice format. Chi-squared, Fisher's exact, and univariable logistic regression tests were employed to assess performance differences in relation to question characteristics. Results: ChatGPT (GPT-3.5) and GPT-4 achieved scores of 73.4% (95% confidence interval [CI]: 69.3-77.2%) and 83.4% (95% CI: 79.8-86.5%), respectively, relative to the user average of 73.7% (95% CI: 69.6-77.5%). Question bank users and both LLMs exceeded last year's passing threshold of 69%. While scores between ChatGPT and question bank users were equivalent (P=0.963), GPT-4 outperformed both (both P<0.001). GPT-4 answered every question answered correctly by ChatGPT and 37.6% (50/133) of remaining incorrect questions correctly. Among twelve question categories, GPT-4 significantly outperformed users in each, but performed comparably to ChatGPT in three (Functional, Other General, and Spine) and outperformed both users and ChatGPT for Tumor questions. Increased word count (odds ratio [OR]=0.89 of answering a question correctly per +10 words) and higher-order problem-solving (OR=0.40, P=0.009) were associated with lower accuracy for ChatGPT, but not for GPT-4 (both P>0.005). Multimodal input was not available at the time of this study so, on questions with image content, ChatGPT and GPT-4 answered 49.5% and 56.8% of questions correctly based upon contextual context clues alone. Conclusion: LLMs achieved passing scores on a mock 500-question neurosurgical written board examination, with GPT-4 significantly outperforming ChatGPT.

show abstract

On Chatbots and Generative Artificial Intelligence

Cited by 22 publications

References 6 publications

Digital Information Literacy Skills among Library and Information Science Professionals in University Libraries of Sindh Pakistan

Digital Information Literacy Skills among Library and Information Science Professionals in University Libraries of Sindh Pakistan

ChatGPT Needs SPADE (Sustainability, PrivAcy, Digital divide, and Ethics) Evaluation: A Review

Performance of ChatGPT and GPT-4 on Neurosurgery Written Board Examinations

Contact Info

Product

Resources

About