Research Note: The GartnorGroup A.I. Stack
5-sentence paragraphs for each level of the AI stack based on industry predictions
Base Load Power Supply
The massive adoption of GenAI across 90% of enterprises by 2025 will create unprecedented demands on power infrastructure. Supporting this scale of AI deployment, especially for computationally intensive multimodal applications, will require significant expansion of reliable and sustainable power sources. The expected 30% project failure rate by 2025 could be partially attributed to organizations not adequately planning for or securing sufficient power resources. The growth of multimodal AI from 1% to 40% by 2027 will further intensify power demands as these systems typically require more computational resources than single-mode AI. Organizations must consider power infrastructure as a critical foundation of their AI strategy, not just an operational detail.
Resource Access
The predicted rapid expansion of GenAI adoption will create intense competition for critical AI-enabling resources such as specialized processors, memory components, and networking infrastructure. With 80% of enterprises expected to deploy GenAI APIs/models by 2026, organizations must secure reliable access to these resources well in advance. The 30% project failure rate by 2025 suggests that many organizations are underestimating the importance of resource planning and access in their AI initiatives. The shift toward multimodal AI (growing to 40% by 2027) will further strain resource availability as these systems typically require more diverse and sophisticated components. Successful organizations will need to develop comprehensive resource access strategies that include redundancy plans and strategic partnerships with key suppliers.
Chip Architecture and Hardware
The evolution toward multimodal AI solutions, growing from 1% to 40% by 2027, will drive significant innovations in chip architecture to handle diverse data types efficiently. Supporting 90% enterprise adoption by 2025 will require massive scaling of specialized AI hardware production and distribution. The predicted 30% project failure rate suggests that many organizations may struggle with hardware selection and integration challenges. The requirement for 80% of engineering workforce to upskill by 2027 includes significant hardware architecture knowledge to effectively deploy and maintain AI systems. Organizations must carefully balance their hardware investments with their AI ambitions, ensuring they have the right infrastructure to support their planned use cases.
Components and Devices
The push toward 80% enterprise adoption of GenAI APIs/models by 2026 will drive unprecedented demand for specialized AI components and devices. The growth in multimodal AI capabilities will require more sophisticated and diverse component integration to handle various data types effectively. The high project failure rate (30% by 2025) indicates that many organizations may struggle with component selection, integration, and maintenance. The massive upskilling requirement (80% of engineering workforce by 2027) reflects the growing complexity of AI component ecosystems. Organizations must develop deep expertise in AI components and devices to ensure successful long-term deployment.
Sensors, Signals, and Signatures
The growth in multimodal AI to 40% by 2027 requires sophisticated sensor integration and signal processing capabilities. The prediction that 90% of companies will use GenAI as a workforce partner by 2025 demands reliable sensor systems for effective human-AI interaction. The high project failure rate (30%) suggests many organizations struggle with sensor integration and signal processing. The massive upskilling requirement (80% of engineering workforce) includes significant sensor and signal processing expertise. Organizations must develop comprehensive sensor strategies to support their AI initiatives effectively.
Networking and Cybersecurity
The projection that 90% of companies will use GenAI as a workforce partner by 2025 creates immense pressure on networking infrastructure and cybersecurity systems. The high rate of project failures (30% by 2025) due to inadequate risk controls highlights the critical importance of robust security frameworks. The increase in multimodal AI adoption to 40% by 2027 will require more sophisticated network architectures to handle diverse data types securely. The need for 80% of the engineering workforce to upskill reflects the growing complexity of AI security requirements. Organizations must prioritize networking and security infrastructure to prevent their AI initiatives from becoming potential vulnerabilities.
Algorithms and Data Structures
With 30% of GenAI projects expected to fail by 2025 due to poor data quality, organizations must invest heavily in robust data structures and algorithms. The transition to multimodal AI (growing to 40% by 2027) will require more sophisticated algorithms capable of processing and integrating multiple data types effectively. The requirement for 80% of engineers to upskill by 2027 heavily emphasizes algorithmic knowledge and data structure expertise. The prediction that 90% of companies will use GenAI as a workforce partner by 2025 necessitates algorithms that can reliably integrate with human workflows. Organizations must develop deep expertise in AI algorithms and data structures to ensure successful deployment and maintenance of their AI systems.
Software Optimization
The prediction that 80% of enterprises will use GenAI APIs/models by 2026 creates an urgent need for highly optimized software infrastructure. The growth in multimodal AI capabilities requires sophisticated software optimization techniques to handle diverse data types efficiently. The high upskilling requirement (80% of engineering workforce by 2027) reflects the growing complexity of AI software optimization. The 30% project failure rate by 2025 suggests many organizations underestimate the importance of software optimization in their AI initiatives. Organizations must invest in software optimization expertise to ensure their AI systems perform efficiently and reliably at scale.
Applications Platform
The prediction that 90% of companies will use GenAI as a workforce partner by 2025 requires robust and scalable application platforms. The expected 30% project failure rate suggests many organizations struggle with platform selection and integration. The growth in multimodal AI to 40% by 2027 demands platforms capable of handling diverse data types and applications. The need for 80% of engineers to upskill reflects the complexity of modern AI platforms. Organizations must carefully select and develop their AI platforms to ensure long-term success and scalability.
Machine Intelligence and Robotics
The growth of multimodal AI from 1% to 40% by 2027 represents a significant evolution in machine intelligence capabilities. The prediction that 90% of companies will use GenAI as a workforce partner by 2025 requires sophisticated machine intelligence systems that can effectively collaborate with humans. The high upskilling requirement (80% of engineering workforce) reflects the growing complexity of AI systems and robotics. The 30% project failure rate suggests many organizations underestimate the challenges of implementing advanced machine intelligence systems. Organizations must develop deep expertise in machine intelligence and robotics to successfully deploy and maintain their AI systems.
UX/UI and Conversations
The prediction that 90% of companies will use GenAI as a workforce partner by 2025 requires sophisticated user interfaces and conversational capabilities. The growth in multimodal AI to 40% by 2027 demands more natural and intuitive ways for humans to interact with AI systems. The high project failure rate (30% by 2025) suggests many organizations struggle with creating effective human-AI interfaces. The requirement for 80% of engineers to upskill reflects the importance of human-centered design in AI systems. Organizations must prioritize UX/UI development to ensure successful adoption and integration of AI systems.
Cryptocurrency and Seignorage
While the predictions don't directly address cryptocurrency and seignorage, the widespread adoption of GenAI (90% by 2025) could create new opportunities for AI-driven financial systems. The growth in multimodal AI capabilities could enable more sophisticated cryptocurrency trading and management systems. The high upskilling requirement (80% of engineering workforce) might include blockchain and cryptocurrency expertise. The 30% project failure rate suggests organizations must carefully evaluate any AI-driven financial initiatives. Organizations should monitor the intersection of AI and cryptocurrency for potential opportunities and risks.