The rapid pace of change in technology continues unabated, with the latest enhancement to artificial intelligence (AI) being agentic AI – a shift from rule-based processes to autonomous decision-making.
For companies to excel in this environment, they need to take a holistic approach as to what is required within their business and industry.
As demand for AI in the workplace surges, this new technology is already transforming as it moves beyond strict, rule-based processes to intelligent, self-directed systems driven by changes in process automation.
Agentic AI is the driving force behind this rapid shift in a technology that has already taken the world by storm and led to a surge – and drop-off – in technology shares in the US as analysts got hooked on the AI hype before having their expectations tempered by progress the market thinks is too slow.
It cannot be disputed, however, that AI is the fastest-growing technology in history, and continues to gain traction at every turn. The latest iteration in this innovative time in our history is that the latest generation of automation − agentic AI − enables AI-powered agents to make decisions, adapt to changing environments and execute complex tasks with limited human intervention.
Unlike the most recent generations of process automation, which follows set workflow processes, agentic AI works autonomously, while continuously learning and refining its actions based on data and context.
AI will need to be operated within carefully designed parameters, guided by relevant legal, regulatory and ethical considerations.
Among some of the real-world applications that will become increasingly prevalent over time are aspects such as AI-powered customer support systems to resolve queries, financial advisory platforms that offer proactive investment recommendations, intelligent robots, as well as virtual assistants seamlessly managing calendars.
Microsoft and LinkedIn have already recorded an increase in the use of AI within workplaces, with their 2024 Work Trend Index finding that, as of last May when the latest report was published, the use of generative AI at work had nearly doubled in the prior six months. Among its conclusions was that three-quarters of knowledge workers use AI at work – based on its survey of 31 000 people across 31 countries.
Gartner predicts that, by 2028, at least 15% of day-to-day work decisions will be made autonomously by AI agents, a sharp rise from the current adoption rate. By the same year, a third of enterprise software applications will include agentic AI, up from less than a percent last year.
Adoption of AI varies from sector to sector, with those in financial services and law mostly using such tools for contract analysis, fraud detection and regulatory compliance amid concerns over AI bias and decision transparency. The manufacturing and logistics industries see agentic AI as a good way to predict maintenance, optimise supply chains and improve demand forecasting.
There will be an increasing need for professionals to design, manage and oversee these autonomous systems; skills that still need to be developed in this nascent age of AI.
In particular, the software engineer of the future will need to be proficient in the latest advancements in process automation, specifically relating to analysis, design and development. These individuals will need to couple a deep understanding of machine learning, natural language processing and large language models, which together, serve as the foundation of AI reasoning and decision-making.
The software engineer of the future – who will undoubtedly need an entirely new title – will have to understand and manage data from different sources used within processes. Infrastructure engineering from a hardware perspective will also be vital, as these new career opportunities will require that professionals be able to develop interfaces between man and machine.
Beyond technical AI ability, prompt engineering and context-aware AI development are emerging as valuable skills. Since agentic AI systems rely on contextual inputs to execute tasks, professionals must learn how to structure prompts and design workflows that enable AI agents to operate efficiently.
However, these skills require a solid governance framework from within which companies and developers can operate, as there are already serious concerns about the potential ethical breaches and non-compliance − especially in an environment where the innovation is advancing at a rapid pace.
Companies will have to ensure that any AI system they implement operates transparently and responsibly, which includes ensuring adherence to personal information protection laws, such as South Africa’s Protection of Personal Information Act.
AI will need to be operated within carefully designed parameters, guided by relevant legal, regulatory and ethical considerations – even if specific legal frameworks have yet to catch up to the technology. This distinct aspect of AI also offers career opportunities.
To fully realise the benefits of agentic AI, businesses will need to have a solid understanding of how this affects their company and current business model, using that process to determine what infrastructure, financial, skills and governance changes need to be updated to ensure compliance with all relevant regulatory requirements.
Behind all these requirements, however, are people. Without adequate change management and skills development, the very implementation of AI will fail to be a business benefit, as it will not achieve support from everyone in an organisation.
The key to success lies in adopting a collaborative approach, through which AI is an intelligent assistant, enabling businesses to benefit from new efficiencies, enhance decision-making and create smarter, more adaptive workflows. By being strategically prepared to embrace this future, businesses can ensure they remain competitive in an ever-changing world.
Share