6 generative AI business myths that will make you rethink everything

The boost to productivity and employee satisfaction is significant, but the next part is where the real ROI will come to help people work smarter and focus on higher value efforts. That same generative AI model — with a little refinement — can fill out other forms for other uses like procurement, accounts payable and requests for proposals. It can also provide Q&A access to data and insights in other functions, including finance, compliance, HR and supply chain management.

In other words, generative AI not only offers a performance uptick wherever it’s deployed. It’s also so scalable that you can implement it throughout your company relatively quickly. The resulting boom in productivity can be so significant that it can boost short-term profits while enabling new business models — such as ones based on hyper personalization for every customer, large and small — that had previously been too costly to be viable.

This potential can’t be ignored. At PwC, our network of firms is already reimagining how we run our business and helping clients do the same. Our US firm, for example, is investing $1 billion to expand and scale our AI capabilities. These new capabilities include a “generative AI factory” to deploy this technology quickly and securely, and a firm-wide generative AI assistant, ChatPwC. PwC Canada, PwC Germany and other territories are similarly going big on generative AI. They recognize it’s a game changer for every industry and region.

GenAI myths that could limit success

One challenge with generative AI comes in getting started. Conventional approaches often don’t apply, and it’s critical to understand what this technology can do. Misconceptions can keep you from taking advantage of precisely those aspects of generative AI that make it so powerful. Here are six of the most common myths that we have encountered — along with more accurate ways to consider this transformative technology.

Myth #1: GenAI won’t affect my business

Like it or not, generative AI will seep into your enterprise. Many employees are likely already experimenting with publicly available generative AI tools. Moreover, generative AI is quickly entering applications, like those for enterprise resource planning or customer relationship management, that you already use to run your business. Google, Microsoft, Salesforce and other software providers are embedding generative AI capabilities in their offerings. Without guidance, governance and procedures to comply with any relevant regulations, your people may not use these new capabilities securely or effectively.

Myth #2: It’s so massive we have to go slow

Given generative AI’s remarkable potential to scale quickly, the effective approach is usually to think big from Day One. Consider how your company might operate, if all your knowledge workers could be 30-40% more productive. Thinking big doesn’t mean that you should try to deploy generative AI everywhere at once. You’ll certainly want to start with a few, carefully selected capabilities. But you’ll want to build a framework — with governance and oversight, tools and skills, cloud capabilities and APIs — that can replicate and scale these capabilities quickly.

Myth #3: Generative AI is too new and risky

Generative AI isn’t all that new. The technology has existed for years. Until recently, it just wasn’t powerful or accessible enough to interest most businesses. And generative AI certainly can pose risks, including potential “hallucinations,” data breaches and legal violations. But generative AI is still AI. A robust responsible AI framework can help manage risks, so long as it covers every stage of the generative AI life cycle from strategy (for the board and CEO) to control (including governance and compliance) and responsible everyday practices that address risks related to cybersecurity, privacy, bias, performance and more. Since generative AI may soon be used throughout your entire organization, it’s crucial to lay the foundations early for safeguards, trusted and ethical use.

Myth #4: Generative AI will replace employees

Generative AI is essentially a prediction tool, trained on past examples of how, for example, human language or software code typically flows. If you give generative AI the right prompt, it can predict what words or snippets of code should come next. That can help generate new material, troubleshoot what people have made and provide faster, easier access to data. But generative AI doesn’t think, and it doesn’t produce true innovation or creativity. What generative AI can do is make your people vastly more productive — if you train and upskill them to take advantage of generative AI and reorganize how work is done. Your people may be very open to this possibility, even excited about it. In PwC’s Global Workforce Hopes and Fears Survey 2023, more than half of the 54,000 employees responding expected AI to impact their career positively over the next five years. Possible benefits cited included increasing their productivity and creating opportunities to learn new skills. 

Myth #5: We’ll need to hire a lot of new talent for generative AI

Generative AI does require specialized roles and skills. These include both existing AI roles, such as data engineer and data scientist, and new ones, such as prompt engineer and model mechanic. But with generative AI, you don’t have to build your own AI models from scratch — as conventional AI usually requires. Instead, most of us will license a private version of a generative AI model that comes pre-built and pre-trained on public data. For greater value, you customize this model with your own data and expertise. This is where your business and domain specialists come in. They’ll be the key to leveraging generative AI for your business and should be trained on new roles and skills. 

Myth #6: We don’t need GenAI for our digital transformation

If you’re working on digital transformation without generative AI’s help, you could be missing opportunities to speed up the transformation and reduce costs. You might, for example, be migrating data from silos to a cloud-native environment, where it will be easier for all your people to use it. Generative AI, with its ability to make sense of unstructured data, can partly automate this migration and then provide easy-to-use Q&A and analytics for your people — even if they’re not data scientists — to make the most of this data. Generative AI can offer similar support — easier access to data and analytics — for other transformation initiatives too, such as those in cybersecurity, finance, supply chain management, tax compliance and marketing.

You have time — but not much

As you might have guessed, we see generative AI at scale not just as valuable, but also as inevitable. But for the next year or two, this technology faces constraints. One is technological: There aren’t enough microchips in the world right now to provide the computing power needed for generative AI everywhere. Another constraint is human: Most workforces haven’t yet been upskilled to use generative AI effectively and responsibly. In a global AI survey conducted by PwC Japan, the top obstacle to generative AI cited by US executives was a lack of knowledge about this technology among existing employees.

These constraints can be overcome. Hardware companies are designing new chips and building new factories. Generative AI model developers are rolling out innovations that can do more with less. And the upskilling of both specialized talent and broader workforces has begun. When this production, innovation and upskilling has advanced enough, workplaces, marketplaces and industries will be transformed. Those companies that fail to adapt will likely be left behind — unable to match their competitors’ productivity and shut out of the new business models that this productivity enables.

That’s why it’s so important not to wait and see, but to think big and act now: Rethink what your company’s knowledge workers can do and how they can do it with generative AI assisting them. Rethink too what your company can do with a more productive workforce and faster access to data and insights. You may, for example, be better able to solve societal challenges, such as those related to sustainability and equity. Then, equipped with this vision and a suitable roadmap, act to bring this technology into your company. With the right approach — one that both manages risks and scales quickly — your company can become a leader in the new world for business that generative AI is creating.