Five Guidelines for Delivering Transformational Value
Unlike traditional AI, a single GenAI model can drive results across multiple tasks across multiple functions and business units. For example, a single GenAI model, properly trained and managed, can help knowledge workers across all areas of your company, including tax, legal, finance, and HR, access, organize, analyze, and act on data. This scalability can deliver a strong return on investment (ROI). And because GenAI can often understand unstructured data, such as phone calls and online activity logs, it can transform activities that traditional AI could not. The result is massive productivity gains that, when combined with new ways of working, make new business models not only possible, but inevitable.
Based on our daily work on AI and GenAI internally and with clients, our ongoing research, and our partnerships with leading AI technology providers, we have identified five guidelines that can help drive both short-term ROI and long-term business model reimagining.
Choose use cases that enable rapid value and scalability
GenAI's most impactful use cases have two things in common: they deliver value quickly and can be quickly scaled across an organization.
Rapid time to value requires a clear value proposition that your data, technology stack, and security environment can deliver. Functional, sector, risk, and technology teams must work together to support it. Scalability is enabled by GenAI's key differentiators. Most GenAI use cases fall into six repeatable patterns:
Consider the “deep retrieval” pattern: train a GenAI model to search for specific information within documents and data. Once you have successfully trained GenAI to extract key terms from customer interactions, you can train the same model to do the same for contracts, tax codes, financial reports, employee resumes, social media posts, and more. This can create exponential value.
Evolving GenAI and data simultaneously
If you and your competitors are using similar GenAI-based models, what gives you an advantage? The answer is data. Use relevant, trusted, compliant, secure and proprietary data to customize your GenAI models with your in-house experience and intellectual property.
You don't have to complete data modernization to start using GenAI – you can proceed in stages, with GenAI unlocking new value from your data at each stage. This approach helps you get stakeholder buy-in for data initiatives that were previously out of reach.
GenAI has another benefit: it can leverage data that may be “trapped” in old strategy documents or customer interactions. Previously, sorting through this data would have required thousands of employee hours. Now, GenAI has partially automated the process, reducing costs and accelerating time to value.
Develop your skills and transform the way you work
With GenAI, you typically don't need to hire many new AI specialists because, unlike traditional AI, you typically don't build your own GenAI models. To achieve risk-managed, high-value outputs using a vendor licensing model, you may need to upskill your workforce, as we did with PwC's My AI initiative. You may also need to cross-train some of your current technology team to help oversee and customize GenAI.
To unlock more value, rethink how you work. What else can your employees do with a GenAI assistant performing menial tasks and providing the data and leading practices to support higher-value work? Consider new ways to drive innovation. GenAI is so accessible that with the right skills and guidelines, anyone in your company can use it to create new products, services, and operational efficiencies.
Accelerate your AI initiatives with Responsible AI
Responsible AI should accelerate AI initiatives, not slow them down. When trust is built into AI from day one, delays and rework are less likely to occur to iron out vulnerabilities or meet new requirements. Greater trust in AI among stakeholders leads to broader buy-in for AI initiatives.
Our Responsible AI Toolkit is built on prepared frameworks, templates and code-based assets. It covers strategy, governance, controls, cybersecurity, upskilling and more. It is designed to reduce bias in AI models, increase trust, enable compliance, secure data and protect privacy.
Importantly, our responsible AI is technology-enabled but human-driven: not only do we provide people with the tools and skills to oversee AI and manage its risks, but we also ensure that informed humans manage the high-risk, high-value decisions involving AI.
Future-proof your AI with an open architecture
GenAI is still in its early stages, and the technology is evolving rapidly. As a result, successful AI initiatives are not a one-and-done endeavor. Instead, you need to be prepared to take advantage of the next innovation.
That's why at PwC, we've adopted a production mindset and platform-agnostic “open architecture” approach for ourselves and our clients. We work across the AI ecosystem, and we encourage our clients to do the same.
