Practical AI Adoption in a Tech Company: A Company-Wide Framework of Guidelines, Use Cases, and Training
Kuo, Chia Hua (2025)
2025
All rights reserved. This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-2025111027557
https://urn.fi/URN:NBN:fi:amk-2025111027557
Tiivistelmä
This thesis focuses on the challenges encountered by a tech company providing software products in the adoption of artificial intelligence (AI) tools. The main purpose of this thesis is to develop a company AI adoption framework that makes it clear to the employees when they can use AI freely, and when it will require review through clear AI guidelines, use cases and training programs.
This thesis used an applied action research approach and qualitative research methods to collect data. The data collection started with interviews, company document analysis, and participant observations to understand the current situation in the company. The analysis demonstrated that there is a lack of shared rules, clear accountability, and basic training for the company's existing AI usage, which limits the safe and effective implementation of AI in different teams. The theoretical framework of the thesis focused on the topics of AI adoption and use cases.
After understanding the current situation and reviewing industry best practice, the AI governance principles, decision-making processes, and ethical guidelines were developed in co-created with the company directors. The company directors participated in these meetings, aiming to address the identified needs and create the framework's elements. Additionally, suggestions from other key stakeholders supported the proposal's development, followed by the validation with the CEO.
The outcome of this thesis is the company-wide AI adoption plan that consists of three elements: (1) the AI guideline with eight principles, four information tags, a scorecard for selecting AI tools and showing ownership, and a "ready to publish" checklist; (2) AI use cases for each team defining the inputs, outputs, safety measures, and completion standard; and (3) two short practical training sessions for each team, and also the office hours for any questions. This framework has been approved by the company CEO and is ready for implementation. The implementation will help the case company in finishing product drafts faster, engaging clearer reviews, communicating more effectively about AI tools, handling rights and personal data with safe and clearer ownership of tools and approvals based on shared guidelines and measurable outcome.
This thesis used an applied action research approach and qualitative research methods to collect data. The data collection started with interviews, company document analysis, and participant observations to understand the current situation in the company. The analysis demonstrated that there is a lack of shared rules, clear accountability, and basic training for the company's existing AI usage, which limits the safe and effective implementation of AI in different teams. The theoretical framework of the thesis focused on the topics of AI adoption and use cases.
After understanding the current situation and reviewing industry best practice, the AI governance principles, decision-making processes, and ethical guidelines were developed in co-created with the company directors. The company directors participated in these meetings, aiming to address the identified needs and create the framework's elements. Additionally, suggestions from other key stakeholders supported the proposal's development, followed by the validation with the CEO.
The outcome of this thesis is the company-wide AI adoption plan that consists of three elements: (1) the AI guideline with eight principles, four information tags, a scorecard for selecting AI tools and showing ownership, and a "ready to publish" checklist; (2) AI use cases for each team defining the inputs, outputs, safety measures, and completion standard; and (3) two short practical training sessions for each team, and also the office hours for any questions. This framework has been approved by the company CEO and is ready for implementation. The implementation will help the case company in finishing product drafts faster, engaging clearer reviews, communicating more effectively about AI tools, handling rights and personal data with safe and clearer ownership of tools and approvals based on shared guidelines and measurable outcome.