A little more than a year ago, AI was widely heralded as an absolute game-changer for business process outsourcing (BPO) and a potential wrecking ball for jobs in the industry.
AI was seen as a universal solution that was poised to revolutionize every aspect of the industry, but this was balanced by concerns that such a powerful technological advance would displace millions of agent employee positions and result in job losses across the industry.
Today, this story has matured. Organizations are looking beyond the hype and taking a pragmatic approach to AI, focusing on targeted, high-impact use cases and recognizing that the true value of AI lies in solving specific business problems, rather than total transformation.
Lessons learned from public adoption of AI
In recent months, several major failures have been made public in the LLM environment, including “meltdowns,” hallucinations, and visibly difficult transitions between versions of AI models. These incidents highlight the risks and limitations of large-scale language models (LLMs) in uncontrolled environments. Attempts to impose stricter guardrails sometimes backfire, revealing new complexities. Although AI continues to develop and improve at an accelerated pace, these setbacks are a reminder that much work remains to develop truly practical and reliable AI at scale.
The visibility of AI’s shortcomings has changed expectations. The industry has learned that AI does not immediately solve long-standing problems and that all deployments must be rigorously tested in real-world conditions. The gap between promises and actual reality is now at the heart of every strategic conversation.
As organizations seek to find value in AI output, unintended consequences become apparent. Overly restrictive controls can stifle innovation and even introduce new risks, as seen in recent efforts by LLMs to manage messaging and engagement. The lesson is that controls must be balanced with flexibility, and human oversight remains essential, especially in high-risk situations such as financial transactions and regulated industries. What is clear is that, certainly for the foreseeable future, AI will enable rather than replace human intelligence.
AI in the enterprise: different situations
Enterprise AI operates in a fundamentally different context. Here, the scope is narrower and the environment more controlled, allowing for reliable and impactful implementations with significant business benefits. In BPO, the greatest value is realized when applying AI to well-defined, repetitive tasks where accuracy, efficiency, and consistency are paramount. Rather than being replaced by AI, human agents will augment their output by carefully automating mechanical and repetitive tasks.
Consider a mundane but important task: the post-call summary. Previously, agents were spending valuable time documenting calls, but this process was fraught with inconsistency and inefficiency. AI-driven automation has revolutionized this workflow, providing accurate, standardized summaries and freeing agents to focus on customer service. The benefits are significant: increased operational efficiency, reduced costs, and improved customer experience. This is a prime example of AI as a focused problem-solving tool rather than a panacea. This workflow puts humans in charge of oversight and quality assurance, and leverages creativity and judgment when automation is insufficient.
AI technologies are increasingly being leveraged to simplify and streamline back-end processes, with Agentic AI becoming a key driver of workflow re-engineering. Although this is invisible to the customer, this automation not only brings tremendous value to the business, but also frees up human agents to provide high-value customer service, delivering tangible benefits to the customer in terms of brand loyalty and a frictionless customer experience.
Data, data everywhere
One of the consequences of introducing AI in the BPO industry is a significant increase in data collection. AI-driven outputs such as call logs, translation tools, sentiment analysis, and many other processes are significantly increasing the accumulation of data where human speed was previously the limiting factor. This is the next big challenge for the industry.
Approaches to data management vary widely across the BPO industry, with some providers adopting mature, privacy-focused practices while others lag behind with unnecessarily large data lakes. As we move forward, ethical considerations around data privacy and governance, as well as regulatory compliance requirements, will require organizations deploying AI to ask themselves important questions about the data they are capturing and how that data will be classified, stored, and used.
There are legitimate concerns about the commoditization of customer data, and the need to distinguish between data that truly improves the customer experience and data that is exploited for commercial gain. For companies to use data responsibly and ethically to truly improve the customer experience, they will need to think differently. From a pure business perspective, the BPO industry will have to deal with this issue in the near future, as retaining too much data, regardless of how it is acquired, poses risks with no benefit.
Employing AI as a powerful targeting tool
The journey from hype to actual implementation has been informative. For BPO leaders, the message is clear. While we embrace the strengths of AI, we must remain mindful of its risks. Success lies in identifying specific problems and applying technology wisely to support human activities. It’s not about “boiling the oceans” or expecting technology to completely replace human activity without oversight. The most powerful AI deployments are those that provide tangible value, build trust, and enhance rather than replace the human element.
Mervyn Pretorius, CCI Global Group CTO and CCI South Africa Senior Vice President of Development Cobus Pretorius
