The recent Goldman Sachs Research report, “GEN AI: TOO MUCH SPEND, TOO LITTLE BENEFIT?”, has received much media attention for its highly critical examination of the transformative potential of generative AI. The report presents several arguments to substantiate its position that Gen AI is overhyped and unlikely to result in significant productivity gains and economic growth in the near-term.

Experts like MIT’s Daron Acemoglu highlight the absence of immediate “killer applications” that could justify the current levels of investment. He and others also cite a significant mismatch between the substantial capital expenditures on AI technologies–particularly by hyperscalers–and the immediate economic benefits.

On the facts alone they are totally correct. However, like LLMs, themselves, they can compile voluminous amounts of input and come to the wrong conclusion.

Investors’ Interests Differ from Practitioners

It’s important to note that the Goldman Sachs report is aimed at investors, not business leaders trying to navigate the benefits of GenAI within applications used by and for their employees and customers. Investors need a view of the macro landscape in order to make informed decision about the companies they invest in. They also like to see revenue models for apps or services that generate the basis for calculating “free cash flows” and the “present value of future earnings”.

The capital investment in data centers and expenditures on training by hyperscalers are so substantial that GenAI specialists will need to generate enormous earnings just to approach breaking even. This situation draws parallels to the Dot-Com bubble, a period when companies poured millions into capturing web traffic without a viable business model to convert that traffic into revenue. The risk now is that hyperscalers could overextend themselves in the race towards, not just GenAI, but AGI (Artificial General Intelligence). Pursuit of this all-but-inevitable goal could potentially harm their financial stability. This cautionary note underscores the importance for investors to carefully evaluate the sustainability of these ambitious expenditures in the context of long-term profitability.

CX Leaders Experience Tangible Gains

Shifting the focus from investors to business leaders, particularly those responsible for customer experience (CX), reveals a different set of implications. Daron Acemoglu’s assertion that there have not yet been any “killer applications” from generative AI warrants a closer examination. Just a few years ago, the prospect of computer programs surpassing the Turing Test seemed distant. Yet specialists in “Conversational AI” (primarily Speech Processing, Natural Language Understanding, Machine Learning, and Interaction Analytics) had been tackling the problems associated with speech-enabling IVRs for more than a decade. In late 2022, technologies like ChatGPT have made the Turing Test moot by demonstrating an ability to understand and respond to human queries in a manner that often appears indistinguishable from human interaction.

For CX leaders, the advent of large language models (LLMs) like ChatGPT represents a significant milestone. The capability of these models to engage in human-like conversations, provide accurate and contextually relevant responses, and continuously learn from interactions offers unprecedented opportunities to enhance customer service. While the technology may not always be perfect, its progress is undeniably remarkable and has the potential to transform how businesses interact with their customers. The challenge for CX leaders is to harness these capabilities effectively, integrating them into their customer service strategies to deliver more personalized, efficient, and satisfactory customer experiences.

The Potential of Generative AI in Customer Service

When it comes to customer service, having a virtual agent that can understand natural language and generate human-like responses is arguably a “killer app.” In January 2021, Dan Miller wrote a white paper titled “A Clear Path from IVRs to IVAs.” While it did not anticipate the emergence of chatbots surpassing the Turing Test, it highlighted the need for brands to enhance their technology offerings, moving beyond traditional phone tree options and hold music. Miller advocated for strategies like offering callers the option to receive a callback without losing their place in the queue, or providing a link to a chatbot capable of assisting with their needs. His point was that there incremental changes were already making intelligent automation practical.

With the rise of generative AI, Intelligent Virtual Assistants (IVAs) can now offer even greater benefits to customers. These advanced systems invite users to state their needs in natural language, in the language of their choice, and respond with accurate, personalized answers. They can even fulfill requests through backend integrations. This marks a significant improvement over traditional Interactive Voice Response (IVR) systems, which are often limited to scripted interactions and predefined options.

Admittedly, a generative AI-powered IVA currently costs significantly more than a traditional IVR. However, for high-value clients or specific high-revenue use cases, the investment can be worthwhile. Moreover, even the skeptical Goldman Sachs report acknowledges that the cost of generative AI continues to decrease. As these costs decline, the potential for widespread adoption and enhanced customer service experiences becomes increasingly viable. The IVR to IVA transition is just one of many examples of how brands can leverage Gen AI to take major leaps in improving employee and customer experiences.

Don’t Get Caught Up in the Generative AI Fatigue

It’s easy to become fatigued by the constant buzz surrounding new technologies, and generative AI is no exception. The relentless hype can make even the most enthusiastic individuals feel skeptical and inclined to play the contrarian. While generative AI is currently experiencing this moment of intense scrutiny, it’s crucial to avoid falling into the trap of excessive negativity.

History provides a valuable lesson: after the Dot-Com bubble burst, many companies hesitated to invest in their online presence, only to see competitors who embraced the risk gain a significant advantage. Generative AI represents a similarly transformative opportunity. This technology enables software to comprehend and respond accurately to human language in various forms—transcribing speech to text, summarizing content, and answering questions with remarkable precision. Such capabilities are not just incremental improvements but represent a fundamental shift in how we interact with technology.

Hyperscalers are investing heavily in generative AI because they recognize its potential to revolutionize industries. Being overly cautious or dismissive of this technology could mean missing out on substantial opportunities for innovation and growth. While it’s wise to maintain a realistic perspective, it’s equally important to avoid becoming a naysayer. Embracing generative AI and leveraging its capabilities can provide a significant competitive edge and drive your brand forward in this new technological era.

Categories: Intelligent Assistants