The smart building industry is on the cusp of a significant transformation driven by Agentic AI, a particular type of Artificial Intelligence (AI).
Elisa Rönkä, a former Siemens executive who has spent years working in smart buildings and SaaS and founder of KASVU Growth Hub, believes the industry needs a fundamental shift in mindset to realize the full potential of Agentic AI and move towards truly autonomous buildings.
Foundational issues such as data availability, consistency, and standardization are currently holding back the industry.
The building SaaS market is currently in a state of flux, characterized by being unconsolidated, with many acquisitions and new players. While some SaaS companies have risen to the top, others are not doing well after only two years, indicating a dynamic but unstable market. Despite ongoing discussions around building SaaS applications, the industry has not been able to deliver the desired value or innovation.
According to Rönkä, the core problem is that use cases for smart buildings have remained stagnant for the past 5 to 10 years. Although the industry keeps discussing the same applications, it does not achieve the promised value or see innovation. Rönkä believes the focus should be improving existing use cases rather than chasing new ones.
There is also a significant gap in user experience. Rönkä notes that in the B2B sector, 72% of customers expect personalized content and a consumer-grade experience, and the industry is falling short in this area. Current SaaS offerings are often clunky and far from customers’ expected user-friendly solutions.
Agentic AI paradigm shift
While generative AI has captured the tech world’s imagination, Agentic AI will be the real transformative force in the smart building industry, according to Rönkä. She argues that it will move beyond the limitations of current AI applications and lead to diminishing existing SaaS business models.
Microsoft CEO Satya Nadella believes that business applications will likely collapse in the agent era because they are essentially databases with business logic, and this logic will shift to AI agents. He envisions a future where the AI tier centralizes all logic, allowing for the replacement of backends and creating more AI-native business applications orchestrated by AI agents.
According to Gartner, by 2028, 33% of enterprise software applications will include Agentic AI, up from less than 1% in 2024. This will enable 15% of day-to-day work decisions to be made autonomously.
Rönkä characterizes Agentic AI by its autonomy, adaptability, and social intelligence. She explains that it can make decisions and take actions with minimal human supervision. Furthermore, it learns, adapts, and navigates complex contexts, including interactions with humans and other agents, differentiating it from predefined tasks and workflows. Gartner predicts that the use of Agentic AI in enterprise software will jump from 1% to 30% in five years, further emphasizing that this fast-paced change will transform the industry.
Instead of a simple subscription to software, the industry is moving towards a “service as a software” model. Rönkä clarifies that all or part of the workflow tasks are integrated and embedded into the software, representing the next step toward fully autonomous buildings.
Real-world applications and supply chain
Rönkä envisions that Agentic AI will be embedded in buildings, automating tasks and workflows below the waterline rather than as an additional layer on top of existing systems. She points to several key use cases that will emerge:
- Prescriptive maintenance will be fully automated, from identifying a problem to dispatching service personnel, and a more streamlined and efficient supply chain for replacement parts and maintenance services.
- Agentic AI can identify cleaning needs, e.g., based on expected space utilization, optimize cleaning processes, and dispatch cleaning.
- Digital twins will evolve from simple simulations to complex environments that optimize retrofits, construction, and building operations, leading to more efficient supply chain management for construction materials and building components as AI identifies the best options and quantities needed.
Barriers to adoption
While Agentic AI has tremendous potential, Rönkä stresses the importance of addressing fundamental challenges to realize its full potential. She identifies a significant problem with data availability, consistency, standardization, and interoperability. Rönkä notes that the industry lacks the historical data for AI to learn effectively and is still struggling with basic connectivity. In addition, she says, the perceived importance of data is low, particularly in the absence of regulatory pressure.
Rönkä refers to a tech report from Siemens, her former employer, which highlighted that even when data types are available, executives often don’t see the importance of this data for leveraging intelligent solutions.
Beyond data, Rönkä points to other significant barriers:
- The industry is stuck with a legacy “if it ain’t broke, don’t fix it” mindset, contrasting with the software industry’s view that “if it ain’t broke, it’s obsolete.”
- There is a skills gap in building environments and AI technologies.
- The fragmented SaaS landscape creates a poor user experience and high operational expenditures.
Data-driven mindset shift
Rönkä calls for a shift in the industry’s approach to data to overcome these barriers. She says the industry must start by defining the outcomes it wants to achieve, then determining the necessary data, and finally figuring out how to collect it. She argues that currently, the industry is doing the opposite, collecting data without clearly understanding its purpose.
Rönkä also emphasizes the importance of establishing semantic building domains using frameworks like Haystack, Brick, or Real Estate Core to ensure data consistency. In addition, she urges the industry to think holistically beyond the cloud, noting that a lot of intelligence can be achieved at the device and edge levels.
According to Rönkä, optimizing the basics at the building level is crucial, as well as asking what intelligence can be gained there rather than focusing solely on cloud-based solutions.
Navigating uncertainties
Rönkä acknowledges that external uncertainties could also impact the industry. She points to the deprioritization of ESG topics in North America and investor pull-out from net-zero coalitions, which could impact sustainability-related SaaS businesses.
Additionally, Rönkä notes that the return-to-office mandates may or may not be a technology pull for smart building applications. She says that it remains to be seen whether it creates opportunities to master and innovate workplace applications or if it will be more of an iron fist mandate not driven by technology needs.
The future of smart buildings
According to Rönkä, the future of smart buildings relies on Agentic AI, but the industry must master the basics first. She believes data quality, interoperability, and strategic value must be priorities. Ultimately, Rönkä argues that a fundamental mindset shift is necessary to embrace change, invest in data and skills, and move from siloed SaaS solutions to integrated, service-based offerings driven by Agentic AI.
Rönkä concludes by emphasizing that while the industry’s challenges are real, so is the potential for change and innovation. By focusing on these fundamental shifts, she believes the industry can unlock the transformative power of Agentic AI and move towards genuinely autonomous buildings.
2/10/2025 | Elektrik - Elektronik Mühendisliği
All information, documents, photos, videos, images, texts, etc. on our web pages cannot be used without permission and without citing the source. Otherwise: Legal and criminal action will be taken within the scope of Law No. 5846. All rights of Şırnak University are reserved.