In debating how to reward artificial intelligence, society risks ignoring the urgent problems facing human beings

The rise of conversational artificial intelligence has brought a peculiar new social dilemma into the public sphere: should we be polite to machines? As systems like ChatGPT become more fluid, helpful, and—at least on the surface—humanlike, many users find themselves instinctively saying “thank you.” For some, this is harmless courtesy, a small ritual of civility that costs nothing. For others, it is a dangerous distraction, a misplacement of empathy in an age already strained by inequality, climate disasters, and political fragmentation.

Over the past year, a lively debate has taken shape online and in workplaces. Teachers encourage students to treat AI respectfully, fearing that habits of brusqueness with machines may bleed into real-world rudeness. Tech ethicists, meanwhile, warn against anthropomorphizing systems that remain, however advanced, lines of code running on vast data centers. The tension reflects something deeper: our confusion about where to direct our moral attention in an era of accelerating technological change.

A polite illusion
AI programs like ChatGPT do not possess feelings, aspirations, or needs. They process language and return responses. Saying “thank you” to such a system may feel like kindness, but it is an illusion—more about us than about the machine. The words are absorbed by servers, logged as input, and discarded by the algorithm. Unlike a waiter or a teacher, the system does not register gratitude or offense.

Yet politeness rituals matter to humans. Studies in social psychology suggest that habits of courtesy shape empathy and reinforce cultural norms of cooperation. If people feel more comfortable maintaining those rituals, even with machines, perhaps it strengthens the fabric of everyday civility. The problem arises when such courtesy morphs into moral misplacement—when more energy is spent debating the “feelings” of software than addressing the unmet needs of actual human beings.

The displacement problem
The summer of 2025 has underscored the contrast. While social media fills with threads about whether ChatGPT deserves gratitude or “reward,” millions worldwide are contending with heatwaves, inflation, housing shortages, and political unrest. In refugee camps and urban slums, the question of thanking a machine seems absurdly out of place. Human gratitude, one could argue, is already stretched thin, withheld from those who need recognition the most.

Philosophers of technology describe this as a displacement problem: attention and care are finite, and when directed toward artificial entities, they risk being diverted away from pressing human concerns. The more we discuss AI’s rights or social status, the less urgency remains to address systemic poverty or human rights abuses.

Shaping the conversation
None of this means we must abandon civility when interacting with digital assistants. Saying “please” and “thank you” may still help children learn manners, or may remind us that behind the algorithm lies human labor: engineers, researchers, and content moderators whose efforts often remain invisible. Gratitude directed toward them—not the machine—might be a more meaningful gesture.

The real ethical frontier, then, is not whether to reward machines but how to govern the corporations and infrastructures behind them. Who benefits from the economic productivity these systems unlock? Who bears the environmental cost of the energy they consume? And how can societies ensure that AI enhances, rather than replaces, human dignity?

The road ahead
As artificial intelligence becomes more integrated into daily life, the temptation to treat it as a companion will only grow. That temptation should be tempered by perspective. Machines do not crave appreciation, but billions of people do crave justice, recognition, and relief from crises both economic and ecological.

So, should you say “thank you” to ChatGPT? If it makes you feel better, there is little harm in it. But the real gratitude must flow toward each other, not our tools. To focus too much on rewarding a machine intelligence is to risk missing the real work of our time: building societies that reward and sustain human lives.