Engineers' Perspectives on the Use of Generative Artificial Intelligence Tools in the Workplace

The integration of artificial intelligence (AI) into the workplace requires commitment not only from the leadership of company directors but, equally important, from the engineers responsible for its implementation. This article presents a survey on perceptions concerning the utilization of AI tools in engineering environments. It was focused on engineers and students in the areas of electricity, electronics, and computing. The questionnaire covered demographic information, AI knowledge level, preferred tools, primary applications, perceived impact, and attitudes toward labor substitution. With the endorsement of the IEEE Uruguay Section and IEEE Region 9, the survey was distributed via email to a diverse group of potential participants, in several countries, including all IEEE members in Region 9. There were 375 replies to the survey, from 20 different countries in the Americas.


I. INTRODUCTION
THEl atest advances in generative artificial intelligence (AI) have produced a great social impact and are having a strong impact on work activity in various industries and sectors.Companies are introducing these technologies at all levels,from the management decisions to the production areas and the innovation centers.Some researchers have addressed how new AI techniques can help in prototyping and in the innovation process [1], [2].Others explore how top management can encourage and facilitate AI -enabled business model innovations [3] and the impact of the adoption of AI techniques in industrial production systems and some specific industries, such as manufacturing [4].However,it is still not clear the degree of adoption that these technologies are having in various work environments.Many people believe that AI will replace,at least in part,some work activities.The incorporation of AI into the workplace requires not only the conviction of top management,but also,fundamentally, of the engineers and technicians who must apply it in their daily work.Some surveys and studies have been performed with regard to the impact of AI in the workplace.The Organization for Economic, Cooperation and Development has recently published the results of a survey performed with workers and employers in finance and manufacturing industries [5].The survey reveals that workers in these industries tend to be very positive about the impact of AI on their work.Around 80% of AI users said that AI had improved their performance at work,compared to 8% who said that AI had worsened it.Nevertheless,some workers express some concerns about the impact of AI on job stability and wages.In a general survey regarding how Americans think about AI [6],nearly half (46%) of the responders are very or somewhat excited about AI that could perform repetitive workplace tasks,compared with 26% who would be very or somewhat concerned about that.In a study conducted in 12 Western European countries,it is shown that 67% of European public organizations have adopted one or more AI use cases [7].
These available reports and surveys were performed in some industry or public sectors,or to the general public.Nevertheless,none of them studied the impact of AI specifically in the field of engineering.This motivated our survey,which focused on knowing the level of knowledge, use,and perspectives of engineers and students,mainly associated with IEEE,in relation to AI tools in their work.The degree of knowledge, the used tools,the predominant uses,the impact,and the perceived perspectives regarding labor replacement are investigated and described in the following sections.These results may be taken into account for managers in any industry whose workers are electrical or computing engineers.

II. STUDY METHODOLOGY
This survey was carried out on engineers and students with an electrical,electronic,and computer profile,with the aim of collecting data on their experience and perceptions about the use of AI tools at work.The survey was designed by the authors to obtain demographic information, level of knowledge of AI,tools used, main uses,perceived impact,and attitudes toward labor substitution.
The survey was available between May 4,2023 and July 7,2023,and a total of 375 responses were obtained (corresponding to a 2.7% response rate).In order to protect the privacy of respondents,the responses of the survey were kept anonymous.We took deliberate measures to ensure that no personally identifiable information or individual details were associated with the answers provided by the participants (i.e.,the survey did not collect names,addresses, phone numbers,email addresses, or any other personal information from the respondents).Survey results Figure 1.
Country where you work or study.
Main area of activity.
Authorized licensed use limited to the terms of the applicable license agreement with IEEE.Restrictions apply.
are presented in an aggregated and summarized form.This means that data were collected,analyzed,and reported in a way that prevented identification of participants.
The survey was tested and validated by a group of IEEE Uruguay Section colleagues.After validation,it was distributed via email to a broad audience of potential participants using a recognized IEEE eNotice mailbox,along with the following text: The survey is short and will only take a few minutes of your time.Your response is anonymous.The final results will be published through IEEE channels.
Please click on the following link to access the survey:

Survey on the use of AI
We thank you in advance for your participation in this survey." Two identical mails were sent, the second as a reminder,with a separation of ten days between them, to a total of 13 825 IEEE Region-9 members.
The survey questions are detailed in the Appendix.

III. RESULTS AND ANALYSIS
A total of 375 responses were received from 20 different countries in the Americas.A total of 52% of the responses correspond to professionals who work as engineers  Authorized licensed use limited to the terms of the applicable license agreement with IEEE.Restrictions apply.
or engineering technicians,36% to professionals whose main area of activity is teaching or academia,and the remaining 12% to students.The percentages of responses to each question are shown in the following charts,from Figures 1 to 13.Numbers of the figures are correlated to the question numbers shown in the Appendix.
A. General Knowledge Level and Degree of Use of AI Tools A total of 21% of respondents state that they have a high degree of knowledge of AI tools;44% have a medium degree of knowledge, and the remaining 35% have low or no knowledge.There is not a clear relationship between the degree of knowledge and age.The lowest degree of knowledge is found among those under 30 years old and those over 60 years old:40% of those under 30 years and 41% of those over 60 years have little or no knowledge.
A total of 69% have used an AI tool in the last six months.In this case, 77% of those under 50 years old have used some AI tool,compared to 57% of those over 50 years old.Of the 31% who have not used AI tools,a third indicates that they do not know AI tools.

B. Analysis of AI Tools
Question 8 of the survey asked about the use of several generative AI tools,which are described in Table 1 and the following descriptions.

1) Text Generation Tools
ChatGPT: Chatbot Generative Pre-trained Transformer (GPT) is a prototype dialogue-based AI automaton (chatbot),developed by the OpenAI company [8].Use of any AI application in work in the last 6 months.
If the answer is no, why?
Authorized licensed use limited to the terms of the applicable license agreement with IEEE.Restrictions apply.
of natural language processing, developed by Google [11].
Copy.ai:It is a software that uses AI to generate texts,targeted to business and marketing applications [12].

2) Image Generation Tools
Dall-E: It is an AI system that can create realistic images and art from a natural language description, developed by the company OpeanAI [13].

Stable Diffusion:
It is an AI model capable of generating photorealistic images based on any text input [14].

Midjourney:
It is an AI program with which its users can create images from textual descriptions [15],from the Discord platform [16].Midjourney is an independent research laboratory.

StyleGAN2:
It is an AI model for the generation of face images [17].

3) Video Generation Tools
Synthesia: It is an application that allows to create videos of an avatar speaking,based on textual descriptions [18].

4) Audio Generation Tools
Vall-E: It is a tool capable of simulating voices from a text and a short voice sample,developed by Microsoft [19].

5) Presentations Generation Tools
SlidesAI: It was designed to create professional slides for presentations from any text [20].Authorized licensed use limited to the terms of the applicable license agreement with IEEE.Restrictions apply.
substations and power plants,image recognition,facial recognition,and audio processing,among others.

C. Impact on Work and Future
Perspectives A total of 79% of respondents indicate that the impact of using AI applications has been positive or very positive in their work.Only 0.4% have indicated that the impact of the use of these applications has been negative.It is noteworthy that none of the respondents answered "very negative."Thereare no significant differences in this response according to the age group.
A total of 15% believe that this type of application puts or will put their own work at risk in the near future.This percentage grows to 25% in students.
A total of 76% have an optimistic or very optimistic perception about the impact of AI in their profession.This overall percentage rises to 83% in the case of students and drops to 71% in the case of engineering professionals.A total of 57% have an optimistic vision for the future of other professions.In contrast,there are 7% of respondents who have a pessimistic or very pessimistic view about the future impact of AI in the engineering profession and 17% in other professions.

D. Analysis by Age
The results indicate that,even in a group with technical training in engineering,the degree of general knowledge of AI tools is low in all age groups.A total of 21% of respondents have a high degree of experience in AI tools.The knowledge level is not equally distributed among the professions: 34% of professors or academics reported a high degree of knowledge regarding AI tools,compared to 26% for computer engineers,13% for electrical engineers,and 7% for students.Even so,the degree of use of these new techniques is high:almost 70% of the group surveyed have used some AI tool, recently.Notably,the younger respondents exhibited even greater adoption, with 85% of those under 30 years old actively using AI technologies.
Similarly,academics (students and professors) report 79% of adoption, compared to 59% in the case of professionals.
It is remarkable that students are the segment with lower knowledge in the technology (7%),but with higher adoption of the available AI tools (87%).Only 10% of the total respondents stated that they did not know AI tools.

E. Applications of AI at Work
The most frequent use of AI techniques is associated with the generation and analysis of texts,with ChatGPT being the most used application among respondents,by a wide margin.
Another common use is general information searching.Software code creation ranks fifth in usage.
Overall,the utilization of AI applications at work yielded predominantly positive outcomes.Particularly,noteworthy is the significantly higher positive impact observed among younger respondents,with 86% of those aged under 40 expressing favorable views.Furthermore,within the academic community (professors and students), 81% of respondents believed that the impact of AI on their work was positive or very positive.Even though 15% of those surveyed believe that their job could be replaced by some form of AI,in general,an optimistic view of the future impact of AI applications is perceived.It is noteworthy that the impact of AI is perceived as more positive in engineering-related professions, compared to other professions.Also, optimism decrease as age increase (from 81% for those with less than 30 years to 67% for those with more than 60 years old).It is worth mentioning that students are the segment most apprehensive about the threat AI poses to their future jobs,with 25% expressing such concerns.However, they paradoxically are the most optimistic group regarding AI's impact Figure 10.Use of AI applications that impact work.
Risk of impact of current AI applications on job reductions, in the near future.
Authorized licensed use limited to the terms of the applicable license agreement with IEEE.Restrictions apply.
on their future profession,with 83% expressing positive expectations.

IV. CONCLUSION
In the present work,the level of general knowledge and the degree of use of AI tools among professionals and students of electrical and computer engineering have been analyzed,as well as the most used tools and their specific applications in the workplace.In addition,the impact that these tools generate at work and the future perspectives for their adoption have been examined.This information is useful and should be taken into account in any management decision.
The information was obtained by conducting a survey,to which 375 professionals or students responded,directly related to the areas of electrical engineering or computing,residing in 20 countries in the Americas.
From the analysis of the collected data,the following conclusions have been obtained.
The results of the survey reveal that the level of general knowledge about AI is low.However,there is a great adoption of the use of AI tools in the workplace.Among the most used tools are those related to the generation and correction of texts,where ChatGPT stands out as the most used by a wide margin.Complementarily,many other AI tools are used,mainly for general information search,text generation and review,data analysis, and software code creation,among other functions.
The adoption of AI tools seems to have had a significant impact on the work of the professionals and students surveyed.A very positive perception has been evidenced regarding the impact of the use of AI in the work of professionals.
In general,the future perspectives regarding AI at the workplace are positive,both in the profession of the respondents and in other Figure 12.Perception of the future impact of AI applications in the engineering profession.
Figure 13.Perception about the future impact of AI applications in other professions.
professions.The vast majority of those surveyed are not afraid of being replaced at work by AI techniques.However,a not insignificant number of professionals believe that this will be likely or very likely in the future.
In conclusion,the report shows that the level of knowledge about AI is low,but its use is wide.Text generation tools are the most used in the workplace.The impact at work has been perceived as positive.Future perspectives are mostly optimistic,both in the engineering professions and in other professions.

APPENDIX SURVEY QUESTIONS
The questions asked were the following (presented in Spanish and English).

Figure 4 .
Figure 4.Level of knowledge regarding Generative AI tools.

AutoGPT:
It is an open-source experimental application that uses the capabilities of the GPT-4 language model [10].GPT-4 is the most advanced text generation model from the OpenAI company [9].BERT: Bidirectional Encoder Representations from Transformers (BERT) is a technique based on neural networks for the pretraining

Figure 5 .
Figure 5.Use of any AI application in work in the last 6 months.

Figure 8 .
Figure 8.Specific tasks or activities with the help of AI applications.

Figure 9 .
Figure9.Main specific uses with the help of AI applications

Table 1 . AI Tools Used.
1) Select the country where you work or study.2) Select your main area of activity.3) Select your age range.4) Select your level of knowledge regarding Generative AI tools.5) Have you used any AI applications in your work in the last 6 months?