The advent of AI agents is building capabilities that can overtake today's cloud service software used by the majority of global companies. These intelligent systems act autonomously, directly interfacing with core databases and bypassing outdated workflows, ushering in a new era in operational efficiency that will upend the massive Software-as-a-Service (SaaS) industry that has been relied on for decades.
The SaaS industry was valued at more than $250 billion globally last year, according to US technology consultancy Gartner, and is a cornerstone of how online companies function. However, as industries continue to evolve, the SaaS model – built on static applications layered over databases – is giving way to a transformative paradigm: Agentic AI.
Microsoft chief executive Satya Nadella recently predicted that traditional apps such as e-commerce platform Shopify or project management platforms like Trello or Monday.com, would give way to intelligent agents. This would shake up the foundational framework of the more than 84 per cent of technology companies that depend on SaaS for their operations.
Mr Nadella announced his plan to put AI agents into the mainstream as he introduced Microsoft 365 Copilot Chat in January.
“It's about unleashing a swarm of intelligent agents to supercharge your productivity and unlock the full return on investment in AI," he said in a video.
Similarly, Meta chief executive Mark Zuckerberg has outlined a future where AI agents perform complex tasks, from writing code to optimising workflows.
These advancements signal the dawn of an agentic era, where AI systems directly interact with core databases, bypassing static interfaces and delivering exceptional operational efficiency, precision and adaptability. For businesses, this means rethinking strategies to fully harness these capabilities effectively, with a focus on optimising infrastructure as a key enabler for future AI transformation.
Generative v Agentic AI
Traditional SaaS applications rely on Create, Read, Update and Delete operations, with complex layers of software that intervene between users and databases. Agentic AI fundamentally redefines this dynamic. Rather than static interfaces, these agents act as intelligent orchestrators, autonomously executing tasks and transforming static applications into dynamic, purpose-driven systems. By migrating decision-making and automation to the AI layer, businesses can achieve greater innovation and adaptability.
These breakthroughs are largely driven by advanced transformer-based technology, according to Stanford’s 2024 AI Index, that are enabling the realisation of Agentic AI. Developers face challenges adapting to this design paradigm, but businesses stand to unlock significant opportunities.
The distinction between generative and agentic AI lies in their approach to tasks and decision-making. Generative AI powers popular tools like ChatGPT and Google Gemini, but their capabilities remain reactive, relying on user prompts. In contrast, agentic AI operates autonomously, setting goals, strategising, and adapting in real-time. While generative AI serves as a skilled assistant, agentic AI functions as an independent collaborator, managing workflows, making decisions, and driving outcomes without constant supervision.
For example, generative AI may draft an email upon request, while agentic AI would proactively monitor the inbox, prioritise messages, draft responses, and schedule follow-ups – all while learning and adapting to user preferences.
Real-world applications and implications
Agentic AI holds transformative potential across industries by redefining operational efficiency and delivering intelligent solutions. In customer support, it moves beyond static, script-driven chatbots to dynamically adapt to tone, context, and subtle cues, offering empathetic and personalised interactions. For instance, an AI agent addressing a product availability query can check inventory, consider the customer’s location, and suggest the nearest store or delivery timeline. It even recalls prior interactions to provide a seamless and tailored experience, allowing human agents to focus on complex, strategic challenges.
Similarly in manufacturing, Agentic AI goes beyond programmed routines by actively optimising production lines in real-time, responding dynamically to challenges, and enhancing overall efficiency. Additionally, in workflow management, these AI agents anticipate bottlenecks, suggest process improvements, and autonomously handle tasks, ensuring streamlined operations and maximising productivity. By enabling adaptability and precision across industries, Agentic AI transforms how businesses operate and innovate.
With AI agents poised to transform operational workflows by redefining how data is managed and productivity is achieved, these agents will act as engines of enhanced performance, mediating digital interactions and evolving as repositories of human knowledge. As businesses identify new AI use cases across industries, ethical frameworks and governing will become critical to ensure responsible AI deployment.
Companies must evaluate their readiness by assessing talent, processes, technology, and partnerships, emphasising foundational processes to train, test, and optimise AI systems. By prioritising business processes over technology, organisations can unlock AI’s full potential, achieving transformative outcomes.
As the SaaS era draws to a close, agentic AI emerges as the defining force shaping the future of business technology. Businesses that invest in assessing their readiness across talent, infrastructure, processes, technology, and partnerships will unlock transformative potential. The question is no longer whether businesses will adopt Agentic AI but how soon they can harness its limitless possibilities to reap its rewards.
Trevor North is the Chief Operating Officer at Core42
