The next phase of AI – five key perspectives
08 | 2025 Jari Huilla, Kipinä CTO & Partner
Artificial intelligence is exceptionally prominent in both heated technology discussions and corporate strategies. Consequently, AI strategies are playing an increasingly critical role in companies' future visions. Few topics are as promising yet as widely misunderstood. Over the summer, Kipinä's experts explored five different perspectives in a blog series on what developing next-generation AI solutions truly requires.
One of the key observations was that building AI does not start with code, but with the question of where it is genuinely applicable and what problem it solves, as discussed by Kipinä CTO Jari Huilla in his blog. Applying AI to every conceivable scenario sounds tempting to many, but in reality, the best results are achieved with carefully scoped and well-understood problems. However, this understanding requires the ability to look under the hood, from model architectures to their mathematical constraints.
High-quality, structurally clear, and well-managed data is the cornerstone of AI and must be addressed right at the design table rather than patched on afterwards, points out Kipinä Partner and Data Architect Antti Rintala. A great AI solution always starts with great data is the foundational principle upon which next-generation AI solutions are built. Only sound foundations can create scalable solutions that stand the test of time.
The third insight comes from the PlanetAI case study, where Kipinä's development team helped bring an ambitious vision to life as a production-ready service that helps global enterprises verify their sustainability efforts. The project demonstrated that a successful AI solution requires technological excellence, business acumen, and the ability to build a solution that scales and meets international standards.
Despite nearly limitless possibilities, technology never exists in a vacuum. Cybersecurity and accountability are integral parts of the design and deployment of every AI solution, emphasizes Senior Software & Cloud Engineer Erkka Korpi. The more systems are automated and interconnected, the more crucial it is for humans to maintain control – and to understand when to pause and ask: "As developers and clients, are we sufficiently aware of the ongoing race between AI and cybersecurity in software development? Do we recognize the risks of haste and ever-increasing speed?"
When discussing artificial intelligence, its future, and its possibilities, we cannot overlook humanity and the role of empathy. Psychological safety, transparency, and the ability to view one's own decisions from multiple perspectives are prerequisites for AI to genuinely support collaborative thinking rather than merely accelerating existing workflows. AI can be a powerful tool to support thinking and decision-making, but its potential falls short if we do not dare to be incomplete and ask difficult questions.
By combining these five key perspectives, a comprehensive picture of the "second phase" of artificial intelligence emerges:
Precise problem definition
Robust data foundations
Business-driven implementation
Information security and responsibility
Human-centricity alongside technology
Developing artificial intelligence is not just about the latest technology. It is about the ability to build solutions that are simultaneously efficient, secure, and meaningful. It is a journey that begins with questions—and continues with the courage to examine both data and ourselves more closely.