Artificial intelligence can summarize research, generate ideas and analyze vast amounts of data in seconds. But can it help us make wiser decisions? In an interview with Greater Good at the University of California, Berkeley, Cheryl Strauss Einhorn, author of The Human Edge: Smarter Decisions in the Age of AI, argues that the rise of AI makes certain human qualities even more important: reflection, empathy, values and ethical judgment.
Einhorn is clear: AI has no values. It has biases and influences from those who designed it and from the information fed into its system. That is why people must bring values such as kindness and care to its use. Before we sit down in front of an AI tool, we need to ask ourselves what problem we are solving, why we are solving it and what the context is.
If we let the machine think for us, we may lose our sense of right and wrong, because only we know the details about the people with whom and for whom we are solving a problem. A useful question to put to ourselves or the tool is: before I act, help me think about the people, relationships and responsibilities this decision affects, not just the outcome.
An example from the interview shows the difference. Felice, a food scientist at a pet food company, used AI to analyze her team’s workload and expertise. The system suggested assigning a critical stability test to the most experienced female technician, but Felice understood deeper dynamics: who works best under pressure, who loves innovation and who needs guidance. She ultimately gave the work to a younger scientist who wanted to develop skills, securing both short-term success and long-term growth for the team.
Einhorn uses the cheetah metaphor to discuss the strategic pause. A cheetah goes from 0 to 60 miles per hour in about three seconds, but its hunting ability comes from its capacity to slow down by as much as nine miles per hour in a single stride. This pause creates flexibility and the ability to change direction. It allows us to break learning into chunks and make cognitive space for new information. As she says, there is a difference between speed and quality, and between being effective and being efficient.
She warns that more information does not necessarily mean better decisions. Francis Bacon said that “knowledge is power,” but today we often already have all the answers. Without a clear definition of the problem, motivation and context, we risk analysis paralysis. Einhorn suggests the “vision of success” question: imagine that the decision has already been made well and identify the few things that, if they do not succeed, mean the decision fails. These become the boundaries we communicate to AI so that its research is focused.
The example of Max, who was planning a vacation, illustrates this clearly. Asking AI directly for help, he received destinations, flights, attractions, ratings and price comparisons, but felt paralyzed by the volume of choices. When he used the “vision of success,” he was able to tell the tool: “I want to return home truly rested, having spent quality time with my family, without spending too much.” AI then had a meaningful problem to solve. It cannot define success, but it can optimize for the success we define.
The AREA method, developed by Einhorn, is a framework for challenging assumptions and errors in thinking while incorporating other people’s perspectives. The first “A” stands for Absolute: original, unfiltered information from primary sources. The “R” stands for Relative: outside perspectives and secondary sources. The “E” stands for Exploration and Exploitation: exploration broadens the research through interviews and unusual sources, while exploitation turns inward and challenges personal biases. The final “A” stands for Analysis: it brings everything together and provides clarity in complex decisions.
The method becomes even more important with AI because AI does not eliminate bias; it can reflect it and sometimes amplify it through training data, design and the way it is used. Einhorn notes that much of AI’s information comes from Reddit, may be incomplete or outdated, and that the system sometimes gives wrong answers. That is why a process of checking is needed, rather than simply accepting the answer. AREA helps make the use of AI a conversation rather than a transaction, with a perspective grounded in evidence and different stakeholders.
The new competitive advantage, according to Einhorn, will be self-awareness. AI is the first tool that asks for part of our cognitive load and invites us to hand over decisions. If we hand over that load, our thinking weakens; but if we understand our metacognition, we can strengthen our thinking tools. The invitation is to turn to ourselves first: to ask why we are using the tool and whether we should turn to ourselves instead. This allows us to solve problems together with AI, but also beyond it.
Comments