Transformative Progress in 2024

The year 2024 was a watershed moment for technological advancements, with generative AI taking center stage and redefining personal productivity applications. This year decisively proved that generative AI is far more than just hype. According to NTT DATA’s 2024 global cross-industry survey, only 1% of C-suite executives consider GenAI overhyped; the majority recognize it as a long-term game changer that requires time to optimize. Two-thirds of these leaders view GenAI as revolutionary, though non-executive leaders remain more skeptical, often seeing it as complementary and nonessential for now, or struggling to justify its ROI. This conclusion is not unexpected, as increased hands-on experience within an organization typically brings to light additional hurdles and challenges, such as data availability, legacy infrastructure, and scalability, that relevant positions need to address.

In the industrial sector, Industrial AI continued to reshape operational processes, boosting efficiency, accelerating robot programming, and refining application development through co-piloting engineers. Despite persistent challenges—like sparse data, inconsistent quality, trust issues, and difficulties in transferring best practices across factories—solutions are beginning to emerge. Synthetic data and reinforcement learning are tackling data scarcity, while pre-trained models are making AI more accessible to non-experts.

Yet, 2024 wasn’t solely defined by AI. Cloud technology also saw significant growth, with Gartner forecasting a 20.4% increase in public cloud spending, reaching $675.4 billion by year-end. This growth, driven by generative AI and application modernization, underscores cloud technology's pivotal role in boosting productivity and fostering innovation, setting the stage for 2025.

As organizations in services and manufacturing continue to explore automation and robotics, 2024 offered a glimpse into the future with the rise of human-like robots capable of learning by observing human tasks. While large-scale industrial deployment remains on the horizon, Goldman Sachs Research highlighted a faster-than-expected decline in the manufacturing costs of humanoid robots, with costs dropping by 40%, outpacing the anticipated 15-20% annual reduction.

The driving forces behind these emerging tech trends in the industrial world are clear: the urgent need to improve efficiency, reduce costs, foster innovation, ensure sustainability, and maintain global competitiveness. While areas like quantum computing hold promise, their practical application in industrial settings may still be years away. Instead, the rapid acceleration of AI-driven innovation, particularly in industrial AI, has introduced a broad spectrum of techniques and approaches that demand attention in any forward-looking analysis.

Selected Digital Tech Areas to Watch in 2025

1. AI Shifting from B2C to B2B

AI innovation is increasingly focusing on business-oriented solutions, emphasizing efficiency, automation, and optimization. While consumer-facing (B2C) AI applications, driven by internet-scraped data, continue to thrive, business-to-business (B2B) AI is leveraging time-series data from sensors and control systems, along with synthetic data. In 2025, the search for high-ROI use cases will intensify as companies prioritize resilience, competitiveness, and innovation through data optimization.

2. Copilots and Agentic AI

The AI copilots (assistive tools) and agentic AI (autonomous systems) are shaping the future of human-AI collaboration. Agentic AI combines large language models (LLMs) with traditional programming, offering flexibility and precision for autonomous task execution while maintaining reliability in critical applications like security and calculations.

Tech giants like AWS, Microsoft, and Google are heavily investing in agentic AI frameworks. Amazon’s Bedrock enables multi-agent collaboration for complex workflows, Microsoft’s Semantic Kernel enhances enterprise-readiness, and Google’s Gemini 2.0 advances multimodal capabilities and tool integration.

3. Industrial AI and Data Contextualization

Industrial AI is moving beyond traditional analytics toward deeper data contextualization, empowering smarter decision-making and predictive maintenance. Advanced tools like causal algorithms and object-centric data storage are uncovering direct and indirect explanations for events, revealing intricate cause-and-effect relationships.

According to Dr. Michael Haft, CEO & Founder Xplain Data, “understanding cause and effect based on real-world data - ‘Causal Discovery’ or Causal AI- will soon be one of the mega-topics of AI.”

As data grows more complex and regulations around data protection become stricter, data safety is emerging as a competitive differentiator. Vendors offering advanced, user-friendly AI tools that ensure compliance while delivering tangible business value will lead the market.

4. Leveraging Small Language Models (SLMs)

Domain-specific small language models (SLMs) are gaining traction for their cost-effectiveness and tailored functionality. By narrowing their scope, SLMs address specialized questions more effectively than large general-purpose models.

For example, in investment management, smaller, domain-focused agents trained on proprietary data are enhancing workflows. Leveraging domain expertise and targeted prompt engineering will drive the adoption of SLMs, offering businesses affordability and efficiency.

5. Federated AI Learning

Federated learning is becoming a cornerstone of privacy-preserving AI, enabling collaborative model training across decentralized datasets while ensuring data security and regulatory compliance. The industry experts predict that by 2026, around 10% of global manufacturers will adopt federated learning to scale industrial AI, improving access to the right data, reducing data preparation efforts and enhancing security. Based on real-world results, cost-effective scaling of AI became a new key use case for federated learning in 2024.

This decentralized approach minimizes risks associated with centralized repositories. For instance, Siemens and Katulu have deployed federated learning across multiple factories, improving quality control processes by 80% while reducing cost of scaling and safeguarding data privacy.

6. Human-like Robots

The global humanoid robot market is projected to reach $38 billion by 2035, with manufacturing costs having decreased by 40% in recent years, according to Goldman Sachs Research. China alone showcased 27 humanoid robot models at the 2024 World Robot Conference, highlighting their expanding role in automation, service industries, and human-machine interaction.

Industries are cautiously exploring humanoid robots’ potential. For instance, BMW’s trials of the Figure 02 robot signify the transition from prototype to industrial application. While challenges remain, the momentum toward mass commercialization is accelerating, with exponential growth expected in the coming years.

Additionally, 2025 will bring further advancements in the industrial robotics sector. As Dennis Schmidt-Bordemann, Senior Consultant for Thought Leadership at Siemens, states, " In the short term, we might be surprised by advancements in making ‘old-school’ industrial robots more flexible. For example, AI-powered picking solutions, trained on synthetic data, enable companies to quickly update and enhance entire robot fleets – making them more adaptable and capable in ways that are practical and immediately impactful”

7. The Industrial Metaverse

The Industrial Metaverse is making significant strides in the engineering domain, driving advancements in sophisticated digital twins, immersive simulations, and real-time cross-domain collaboration. This evolution transcends standalone digital twins, envisioning interconnected virtual ecosystems that seamlessly link OEMs, suppliers, customers, and service providers.

Key industry shifts, such as Siemens' acquisition of Altair Engineering, highlight this transformation. It will be intriguing to see how this acquisition enhances AI-driven design and simulation capabilities, bolsters the foundation for engineering digital twins, and accelerates the growth of the Industrial Metaverse.

The world is already witnessing significant deployments of the Industrial Metaverse. For example, HD Hyundai, a global leader in sustainable ship manufacturing, leverages AI-powered, photorealistic digital twins to interactively unify and visualize extensive engineering datasets. Another compelling instance is the immersive industrial metaverse experience at FREYR's Gigafactory, where the physical and digital worlds seamlessly converge.

Conclusion

Looking ahead to 2025, the fusion of AI, automation, and cloud technology is set to revolutionize productivity and collaboration across industries. As Boris Scharinger, AI Strategist at Siemens Digital Industries, puts it, 'I can picture further advancements of GenAI-driven API connectivity, so robotic process automation and, more generally, workflow automation will see the first true benefits in very flexible workflow adaptation.'

From the emergence of domain-specific language models to the adoption of AI-related tools and the evolution of the Industrial Metaverse, the upcoming year promises transformative innovations that will redefine global industrial competitiveness.