Artificial intelligence and emotional awareness
AI will soon be able to recognize, interpret, process, and simulate human emotions. This development naturally raises concerns among users whose emotional data may be analyzed, particularly regarding potential privacy violations.
It is therefore essential that companies begin questioning whether their value proposition aligns with this evolution in technology — whether customers will consent to having their emotions analyzed, and what risks might arise if such systems were to fail.
Leaders should also be aware of the three most common current use cases for emotional AI:
- Systems that use emotion analysis to tailor responses to customers.
- Systems that perform emotional analysis on a specific audience for research and learning purposes.
- Systems capable of imitating (and potentially replacing) human-to-human interaction.
Organizations that manage to anticipate the shift, overcoming the challenges mentioned above while seizing emerging opportunities, will be able to evolve their services and guide the transition from data-driven to emotion-driven customer interactions (deeper, more personal, and more meaningful). The brand itself may be the greatest beneficiary of this paradigm shift.
How AI will transform corporate strategy
AI enables organizations to make more accurate and cost-effective predictions, and once a certain threshold of precision is achieved, this capability will profoundly reshape corporate planning.
Strategic decision-makers should begin by assessing how much these predictions are likely to improve (both in reliability and in speed) according to their specific industry context. They should then consider what new strategic options could emerge in this transformed environment.
Better predictions will foster the creation of new business models, redefining the rules across multiple industries and prompting the need for investments in new types of expertise beyond traditional skill sets.
For instance, an online retailer like Amazon could implement a model based on predictive accuracy (one that might be called “ship-then-shop”) where goods are shipped to customers even before they have selected or purchased them. Amazon might even launch such a system before it becomes profitable, simply to capture the first-mover advantage (a decision that other players in different sectors might emulate to build barriers to entry for competitors). Of course, this strategy would require additional infrastructure investments to manage the inevitable increase in product returns.
The AI of the future will require less data
As AI technologies evolve and spread, they will likely require fewer large datasets. Instead of relying on bottom-up “big data,” they will increasingly simulate human top-down reasoning, approaching problems and tasks through conceptual understanding rather than pure pattern recognition.
This shift will allow for broader applications of AI, creating opportunities — including business ones — in domains previously considered incompatible with such technology.
Until now, AI has advanced mainly through deep learning and machine learning, powered by vast amounts of data to “train” models. However, this approach has exposed significant limitations in scenarios where only limited information is available.
To prepare for the next evolution of AI, companies should plan their investments and experiments in the following directions:
- More efficient reasoning machines. It will be vital to develop robots and systems capable of perceiving and understanding the world around them, allowing for training with far less data.
- Ready-to-use expertise. AI systems will increasingly emulate human experts, operating effectively even in conditions of uncertainty and limited information. The top-down approach could therefore become much more efficient than current data-intensive models.
- Common sense. Top-down AI will enable machines to interpret the world by understanding everyday objects and actions, ensuring proper responses to unexpected situations and learning through experience.
Improved decision-making. Humans often act under uncertainty, basing their decisions on expected value and probability, even with little prior experience. Machines are gradually learning to emulate this reasoning process. (Harvard Business Review, 2019)
