Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Ziva Hallaji is Vice President, Global Head of Salesforce, at Wipro. She is a recognized thought leader in the SaaS and CRM spaces.
Artificial intelligence (AI) is poised to exponentially change the way businesses interact with, support and service customers. In fact, it already is. By implementing AI agents into enterprise customer relationship management systems (CRMs), organizations are beginning to expand CRM capabilities to handle the entire chain of support and service tasks traditionally handled by the call center workforce.
Large enterprise CRM providers like Salesforce, Microsoft and Oracle recently began launching AI agent capabilities that are driving the “agentic” era. These AI agents drive task automation based on specific use cases by adding a large language model layer based on actual interactions with natural language. This allows AI agents to do more than just collect information. Within the system, they look at everything that needs to be done to resolve an issue or inquiry.
For example, an AI agent might be assigned a role that includes performing specific tasks and activities, examining a set of data or environments and taking actions based on this defined role. The task may be simple such as generating and sending an email to customers. Or it may be more advanced such as resolving a support case end-to-end by having a conversation with a customer, leveraging previous case knowledge and triggering a human interaction based on conversation sentiment.
More importantly, AI agents can adjust and learn over time based on their experiences. These agents can iterate on their actions and specific roles, leveraging different data components and make decisions independently. Advanced AI agents have a more specialized focus, can work with other AI agents, execute multistep processes that require judgment and decision-making and mimic human interaction.
As AI agents progress from customer service and support, they will be able to manage deeper day-to-day functions including HR, financial accounting, claim and invoice management, even travel. Eventually, AI agents will offer integrated capabilities that allow consumers to control multi-process functions unlike existing AI that performs individual tasks. Imagine having a single agent that can manage the entire household, from temperature control to starting the robotic vacuum or washing machine, to the selection, ordering and delivery of groceries.
While AI capabilities will improve consumers’ lives, they will also redefine the workforce. With the first wave of AI customer support and service capabilities already in the market, a call center’s personnel might be reduced from hundreds to a mere handful, making it critical for companies and employees to prepare now for this shift. Everyone in the organization needs to take an adapt-or-die attitude, learning how to work with AI as roles change and new skills are needed.
As companies begin to evaluate future talent needs, leaders need to take a fresh look at the business problem and begin defining new career paths that align with AI goals. Decide where you think the market will be in 10 years and what you want to achieve. Map that back to processes, use cases, technology and talent. Essentially build backward by designing processes and technologies that align with your goals and are flexible enough to meet future needs.
How will you address trust and security?
One of the biggest hurdles with AI is the trust and security of data. While there is a seemingly endless amount of information available in the market, finding actual truth is difficult. Since AI gathers data from myriad sources, we must ensure that data is factual. ChatGPT type technologies, for example, aggregate a lot of chatter posted about a topic but lack fact-checking capabilities and often any connection to reality.
The more AI agents are performing the work of humans, the greater the need for trust in the system. How secure is your data, where will it reside and who will have access to it? How will you prevent AI agents from “hallucinating” or creating things that aren’t real? Government regulations are emerging, but organizations need governance frameworks for trust and security around AI technology.
Identifying and validating trustworthy data will create a new market for service providers that provide a trusted source of truth. Companies will pay for data that is certified, validated and fact-checked. This “internet of the future” will take us from an era of information to one of truth. However, the general population must first recognize fact from fiction and demand truth. Given the current plethora of misinformation and the lack of control over it, data trust will likely get much worse before there is a strong demand for truth and movement toward ensuring fact-based information.
Where is your organization in the AI evolution?
When to implement AI agents is on many a CEO’s mind as they balance organizational readiness with the fear of missing out. However, it’s important to not just follow suit and implement AI only to discover your data is not trustable. AI agents need to be trained and tested repeatedly until they’re 100% ready. Launching agents sooner may bring quick financial benefits; but if customers stop using them, millions if not billions in investments may be for naught. CEOs need to think of AI as an evolution that will take years and decades to fully implement instead of expecting quarterly results.
As you begin to build AI capabilities, make sure your front-end AI infrastructure addresses data, security, integration and connectivity. Invest also in the backend infrastructure to allow front-end flexibility as needs, and therefore, applications change. Most importantly, look through a futuristic lens when thinking about how AI can benefit your organization and your customers and how you can best leverage AI to improve your competitive advantage.
Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?
