AI Goldfish or AI Dolphin—Which One Is Your Organization Ready to Be?
“You know what the happiest animal on earth is? It’s a goldfish. It’s got a 10-second memory. Be a goldfish.” It’s a line from the show Ted Lasso, when the irreverent coach encourages a frustrated, overwhelmed, and slightly angry player to choose being a goldfish that can move on quickly. It is an encouragement to find a path through anything that might weigh down your forward progress.
The question being asked here is that when it comes to the application of artificial intelligence, what are organizations ready to do? Are they going to be an AI goldfish, pushing past any past promises, risk, uncertainty, or unrealized impact to embrace today’s vision for artificial intelligence? Or are they going to hold on to those memories like a dolphin?
While a goldfish is said to have a 10-second memory, a dolphin is believed to have a long-term social memory of up to 20 years. Researchers believe this contributes to behaviors like responding in context to past interactions and holding grudges. Now, imagine if all the participants in a digital transformation journey were dolphins.
Taking the same time span of 20 years, it isn’t unreasonable to think that organizations that have embarked on transformative business change programs may have significant experiences that could impact their long-term social memory. It is also not unreasonable to assume that some of those memory scars could include attempts to leverage earlier incarnations of AI, including early machine learning applications that, at the time, promised to revolutionize business and humanity.
Looking back at those expectations for the future of work, most experts believed strides would be taken to drive automations and operational efficiencies. We saw the fast rise of “smart” concepts that reworked aspirations around everything from “smart agriculture” to the dawn of the “smart city.” In the CRM space, we dreamed of a time filled with transaction-driving recommendations and automated digital experiences tied to optimized predictions and real-time engagement. Promises of “personalization at scale” began to fill every vendor pitch. In sales, we saw the rise of AI- and machine learning-guided selling meant to recommend the exact right product the customer needed at the exact right moment.
The words are eerily the same. Personalization. Efficiency. Scale.
If we were dolphins holding a 20-year grudge, there would be little space for AI to unfurl this new era of agentic AI, powered by a new generative model library and access to data that has defied the limitations of data structures. AI dolphins would cling to the familiarity of operations and processes that were tried-and-true, user friendly, and harmless.
We are already seeing signs that organizations are holding on to some of the pain of the past. In new research from Pegasystems, most business leaders expect agentic AI to have a positive impact, with 94 percent of survey respondents indicating they expect AI to have a positive impact on their technology, their customers (93 percent), and their entire organizations (92 percent). The interest is there and could be called enthusiasm, if it were not for the fear-based hesitation that is holding back forward motion.
The study reveals that executives are unsure of taking the leap of AI faith. Moreover, these executives, while excited and interested, are still unclear of what the tangible impact of agentic AI will be for the business, with less than half (45 percent) saying they have a clear or very clear understanding of the benefits.
These same executives admit that they are facing the real danger of being pushed off that ledge. A majority of Pega’s survey respondents (84 percent) say they are facing high or extremely high pressure from the C-suite, the board, and competitors to advance the agentic agenda. In case mounting pressure wasn’t bad enough, there is a growing reality that the chasm they risk being pushed into is growing wider, as only 34 percent admit to feeling ready to implement and manage agentic AI.
If business leaders choose to be AI dolphins, holding on to the fears and uncertainty that the future holds, the risk will increase. AI, and especially agentic AI, is a reality in modern business and technology. Customers are embracing AI-driven experiences while expecting that human assistance, empathy, and connection will still be made available. Moreover, competitors who embrace AI early have likely matured beyond leveraging AI as a tool for efficiency and have progressed to leveraging agentic AI as a tool to achieve decision velocity and exponential growth.
It may be time to be an AI goldfish. Here are three mindset shifts to help let go and embrace goldfish-like AI happiness:
Define agentic AI and swim on. Not understanding what agentic AI is or how it fundamentally differs from other forms of AI is part of the mirage that can stoke uncertainty. Agentic AI is defined as a form of AI that is probabilistic in nature, making decisions and performing tasks independently without needing constant human input or intervention. These AI systems are designed to adapt to conditions, reason to interpret tasks, act via automated workflows, and learn to improve over time. This is not to say you should become an AI expert, reviewing the nuances of AI forms before swimming on; this is to say that defining that swirl that is pulling you below the water is the best way to move beyond it.
Accept that there is no right way to swim. AI, especially where and how organizations take advantage of agentic AI, is more of a rainbow of options than it is a single rigid template. It can, and should, fit the needs and demands of core stakeholders. The real beauty of AI is that thanks to its capacity to ingest, analyze, and normalize complex data sources and types, it can extrapolate far beyond human capacity. It has the capacity to comprehend the entire ocean. AI goldfish can swim happily knowing the ocean is available, but the focus is on the tank.
Experiment with the intent of scaling with goldfish-like speed. Past failings and encounters with risk can put organizations on an endless cycle of isolated experimentation that could unintentionally stall companywide AI aspirations. Establish a clear definition and outline of the business goals and value-driven outcomes the organization expects from AI implementations, then redo that exercise of definition and value expectations by function and process. These definitions can evolve and warrant new tests and experiments, but they should not cause the organization to stop swimming. Stopping, especially stopping with the weight of failings from the past, will lead to sinking. Let the past inform, but not weigh down, how the goldfish swims on.
In the end, perhaps it will be the AI goldfish, in a world populated by AI dolphins, that wins the day, pushing the business forward by being able to shed the weight of the past and persevere in the happiness of trying.
Liz Miller is vice president and principal analyst at Constellation Research, cov ering the broad landscape of customer experience strategy and technologies.