The ethical AI playbook 2.0
”How do we integrate the use of AI into business?”
Improving existing operations and creating new opportunities
While identifying and testing different AI use cases has progressed rapidly in the AECO sector in recent years, the changes brought by AI to companies’ business models have been slower and more experimental. This is natural, as changing existing organizational models or building entirely new ones may require significant investments and changes in operating practices. Taking ethical considerations into account is therefore central to ensuring that AI-related business models can be shaped and developed in a sustainable way. In this section, we examine—while considering ethical practices—the opportunities AI offers for improving existing business operations as well as for leveraging partially or entirely new business opportunities.
Improving existing business operations
How can organisations get started with improving business operations using AI? Business performance can be enhanced with AI, for example, by automating repetitive tasks, analysing large volumes of data, or automating reporting. Typically, an organization’s processes are documented as part of its quality system or enterprise architecture. Reviewing these descriptions is a good starting point for identifying AI-assisted processes, as they provide visibility into human roles and the overall picture of process-generated information. These process descriptions can also be used to assess the potential use-cases of AI agents. Although AI does not replace human expertise or the need for human decision-making, it can significantly enhance efficiency especially as a partner to knowledge workers.
Leverage at least the following perspectives when identifying AI-related business opportunities:
Test AI opportunities in your daily work and start with small experiments that support the growth of personal productivity. This gradually increases the entire organization’s understanding of AI and helps, for example, to better understand its risks. Support experimentation by ensuring the availability of necessary (cybersecure) applications and by training the basics of AI use cases as well as ethical use. Identify and disseminate within the organization solutions that can benefit others performing similar work or support organization-level productivity.
Integrate AI into the organization’s strategic objectives. Ensure that AI investments are aligned with high-level business goals and that the use of AI becomes a systematic process that enables long-term productivity improvement and scalability.
Search for and review your organization’s process descriptions in the enterprise architecture. Examine how workflows and process data are produced and consider whether they could be improved and automated using AI.
Ensure basic data readiness. Ensure that the organization collects and stores sufficiently high-quality data to enable the use of AI. Is data available from key business processes, and is it both consistent and easily usable? It is also possible to get started with incomplete data, as long as its potential biases and inaccuracies are taken into account when interpreting results. Also ensure cybersecure data management.
Utilize existing tools where possible. Before making major new investments, find out whether the tools already in use in the organization include AI functionalities and opportunities to improve productivity. Using familiar tools facilitates getting started and may also make it easier to ensure their cybersecure use.
Collaborate with customers and partners. In addition to improving your own operations, engage in open dialogue with customers and (project) partners about how AI could be used to improve shared activities.
Find a balance between rapid and sustainable progress. The rapid development of AI solutions pushes organizations to move quickly in their own development as well, and waiting for one perfect solution may weaken competitiveness. On the other hand, measured and well-considered action helps prevent risks and supports the organization’s ability to adopt AI use. Proceed incrementally and ensure the balance that works best for you!
Consider whether AI is even the best way to solve a particular business challenge? AI is not always the only or the best way to improve business efficiency—approach its use with appropriate critical thinking in every situation.
Thoughts on the use of AI in design: Better efficiency and creativity can be achieved
One common – and entirely valid – concern is whether the use of AI negatively affects creativity in design work. In design work particular emphasis is placed on understanding AI’s role as an assistant that supports human work without making decisions on behalf of the designer. Ultimately, the designer always has the opportunity – and the responsibility – to evaluate and assess AI-generated outputs as one part of the design process. At its best, AI can also act as an enabler of more creative outcomes, for example by providing additional ideas during the conceptual phase and freeing up time from manual work for more meaningful tasks.
The key point, therefore, is to recognize that AI is one tool among others available to designers, much like building information modelling, for example. AI in itself does not weaken or enhance the creativity or efficiency of the design process, but when used appropriately and with an understanding of its possibilities and limitations, its use can clearly lead to positive impacts.
Thoughts on project management and contracting: AI is a good assistant to a leader, but not a leader itself
From the perspective of a project manager and a contractor, AI offers several opportunities to reduce manual work in everyday project activities. Especially language model–based AI solutions can assist in preparing meeting minutes, risk management, or the analysis of project materials (Nyqvist et al. 2024).
In the project-based work stakeholder commitment and the journey toward the final outcome can often be just as important than the output itself. For example, in schedule planning, a better outcome for the project may emerge through a collaborative and manual process. That commits the parties to implement the schedule—rather than AI solution independently producing an objectively optimal schedule. As in design work, AI is therefore a good assistant, but it does not replace leadership responsibility or human decision-making and commitment.
Thoughts from a real estate business perspective: Leveraging AI is a tool connecting the property lifecycle
From a real estate business perspective, the benefits of AI concretizes in a much longer time horizon than in a construction project. Systematic data collection that begins in building design and construction and continues throughout the in-operation phase enables the utilization of concepts such as smart buildings and digital twins. Through these, for example, a building’s energy consumption or renovation needs can be examined, simulated, and planned more efficiently than before.
On the other hand, over a long-time horizon this data capital may also be lost or become fragmented due to out-dated information systems, changed data management practices, or changes in data ownership. At the same time, it is important to be aware of the risks related to collecting sensitive data about building users, and to ensure that the collection, use, sharing, and storage of this data are carried out ethically. Opportunities related to property valuation and space use management further expand the range of possibilities enabled by AI, provided that the rules for their ethical use are well understood.
Thoughts from the manufacturing industry perspective. AI brings manufacturers, users, and distributors closer together
In the manufacturing industry, AI can be utilized in product quality control as well as in the programming of software used in product manufacturing. In the construction product industry, well-documented processes and operating models provide a solid foundation for the adoption of AI systems. There are significant differences between product categories in the construction product industry, such as building services engineering, electrical products, concrete products, adhesives, and others. For this reason, the ways in which AI is applied vary considerably. For example, in building services engineering, maintenance needs can be assessed and predicted based on AI-driven data analysis, whereas quality control of concrete products can be carried out using image recognition. New business opportunities in product value may emerge almost unnoticed, for instance when the results of AI-based image recognition are used in quality reporting for customers.
AI also brings product users, manufacturers, and distributors closer together. As construction process supply chains become increasingly data-driven through digitalisation, this enables, for example, AI-based product selection for end users. This opens up new opportunities for product manufacturers to provide CO₂ data and other sustainability-related information as part of AI-based product selection.
New business opportunities
Business models in the AECO sector have remained largely unchanged and leveraging AI in business models can also offer significant potential.
In addition to improving existing operations, AI and its utilization can also create new business opportunities. These may include solutions that (i) have been identified earlier but have only become feasible as AI has developed, or that (ii) are entirely new opportunities based on the use of AI.
Business models in the AECO sector have seen little change over recent decades, which in itself raises the threshold for the rapid adoption of entirely new operating models. For example, value-based design work has been discussed for a long time, but in practice the business has largely been based on hourly billing or fixed-price models.
Although developing and deploying entirely new operating models requires considerable effort, creativity, and courage. Promoting them plays a significant role in the long-term development of the AECO sector, alongside point solutions. While the use of AI is often initiated through targeted solutions and the improvement of existing business operations, broader solutions that span the entire value chain and challenge current business models offer greater potential for productivity growth. This highlights the importance of collaboration, for example in building and leveraging shared ecosystems and data platforms. At the same time, AI enable better opportunities for scaling business, which has traditionally been challenging in the AECO sector.
Value-based consulting and data-economy-driven business models are examples of new types of opportunities.
Value-based business models have long been discussed both in general and within the AECO sector, and the potential for their broader adoption is expected to grow as AI becomes more widespread. As routine, labour-intensive tasks are automated, value-based pricing may serve both the client and the consultant better than traditional billing models. Such models could be based partly or entirely on rewarding value creation rather than hours worked, incentivizing service providers to improve the efficiency of their own operations and to focus resources on tasks that generate the highest value.
In addition, the expansion and growing importance of data markets in themselves offer potential business opportunities. Since AI requires high-quality data as its foundation, the emergence of data markets may become part of everyday practice in the AECO sector as well: selling high-quality data or purchasing and utilizing data from others may represent viable business opportunities in the future. Some actors may participate in creating these data markets or in selling data, while others may benefit by purchasing and consuming data available on the market. Proprietary data that contains significant domain-specific expertise could be licensed, or AI solutions built on top of such data could be sold to other actors. In the future, synthetic data is also likely to play an increasingly important role in training AI systems. The use of this “artificial” data can enable more versatile AI training by complementing or expanding existing data foundations, reducing the need to collect new data containing sensitive information and thereby lowering personal data–related risks.
Adapting new business models to one’s own operations also provides step to assess your role in the AI development landscape: do you want to participate as
- a user, focusing on the efficient adoption of generally accepted business models and tools
- as an enabler, offering partners opportunities to succeed, remaining open to new business models and adopting them through your network
- as a frontrunner, bringing new business models to market and paving the way for them as one of the first in the sector; or as some combination of these roles?
Future Outlook
AI is reshaping job roles. Industry 5.0 perspective needs to be taken into account in the future development.
In the near future, the AECO sector is likely to move toward more extensive use of AI, with AI becoming increasingly embedded in core business activities of companies. At the same time, the ethical use of AI will very likely become a baseline assumption for any company that uses AI in its operations or seeks to build new business on AI capabilities.
AI is reshaping current job roles while simultaneously creating entirely new ones. The development of business activities also affects how job roles evolve. Future roles will increasingly emphasize media literacy (including AI-generated media), the ability to critically and constructively assess AI outputs, and an understanding of potential errors and biases in those outputs. As the volume of repetitive tasks decreases, a larger share of working time can be devoted to ideation, critical reflection, decision-making, and learning. For example, Tokoi (2024) suggests that in the future, architects’ roles may increasingly include guiding AI and evaluating the solutions it produces.
With new business opportunities, new job roles are also likely to emerge. These could include roles such as AI systems architect, AI team lead, or AI onboarding specialist. However, the sustainable transformation of job roles also requires strong organizational-level support, for example through training opportunities and the management of cognitive workload, as discussed in the Part 2: Risk management and information security section. At the same time, it is essential to ensure the ethical nature of new business models, such as ensuring that their use does not discriminate against or exploit employees or partner organizations.
Industry 5.0–style development, which focuses on collaboration between new technologies and humans, will increasingly be part of operations in the AECO sector as AI adoption grows. On the one hand, the AECO sector can draw valuable lessons from manufacturing to streamline its own operations, for example by better understanding collaboration between robots and humans. On the other hand, acquiring an understanding of Industry 5.0 and adapting to the changes associated with it may even be a necessity for the sector, as parts of the value chain directly and indirectly related to construction and property management are developing in this direction regardless. At the same time, the importance of ecosystems and collaboration between actors will be further emphasized. Central to this are the open sharing of best practices and the development of shared data platforms, as well as addressing increasingly complex information security considerations.
In relation to an Industry 5.0–style future outlook, the development of different technologies is also likely to become more closely interconnected. Advances in AI support developments in robotics and the use of digital twins, while data collected from construction sites and buildings can be utilized more efficiently than before. As a result, the potential benefits of these technologies increase, but at the same time ethical risks, such as those related to information security and safety, also become more complex.
Consider at least these!
When improving your existing business:
- Test the possibilities of AI in your day-to-day work and start with small experiments that support personal productivity growth.
- Integrate AI into the organization’s strategic objectives.
- Ensure basic data readiness.
- Leverage existing tools whenever possible.
- Collaborate with customers and partners.
- Find a balance between rapid and sustainable progress.
- Consider whether AI is actually the best way to solve a specific business challenge.
- Identify your organization’s process descriptions in the quality management system or enterprise architecture. Review them together with personnel – could any of these activities be automated?
When considering new business opportunities, keep the following in mind
- New business models may include solutions that (i) have already been identified earlier but have only become feasible as a result of advances in AI, or that (ii) are entirely new opportunities based on the use of AI.
- Developing new operating models requires effort, creativity, courage, and often collaboration between organizations, but they can significantly enhance productivity at both the company and industry level.
- Examples of new business opportunities include value-based consulting and various roles in data markets.
Read more
Nyqvist, R., Peltokorpi, A., & Seppänen, O, Integration of generative artificial intelligence across construction management. In IOP Conference Series: Earth and Environmental Science (Vol. 1389, No. 1, p. 012011), IOP Publishing, 2024.
In Finnish: Tokoi Jenni, Tekoäly arkkitehdin työkaluna: tekoälyn hyödyntäminen talonrakennushankkeen arkkitehtisuunnittelussa. Opinnäytetyö, Tampereen ammattikorkeakoulu, 2024, link https://www.theseus.fi/handle/10024/85589,
