There is something uncomfortable about the rise of AI bots in election campaigns, and the discomfort does not lie in the technology itself.

Technology that promises to improve our lives turns out, at the same time, to be a powerful instrument of influence in the battle for public attention. We find ourselves at a crossroads between using technology to ease processes, and the temptation to use that same technology to steer people. Where is the line between influence and manipulation? And what does leadership mean in a time when technology changes not only the playing field, but also the rules of the game?

Technology offers a sense of control in times of political division. That sense is often false.

In times of polarisation and chaos, it feels reassuring when something filters the noise and suggests a clear direction. AI bots can broadcast targeted messages and steer debate, creating the impression that there is a clear answer to the complexity of the world. That gives both politicians and citizens a false certainty about what the right course is.

That technological control can unintentionally narrow the space for genuine debate. Algorithms designed to confirm users mainly present information that matches existing beliefs. What feels like personal preference is in reality often a technology-driven echo of one’s own biases. That mechanism undermines precisely the plurality of opinion a healthy democracy needs to function.

Research into election integrity and AI confirms why this dynamic is so hard to correct once it takes hold. Systems optimised for engagement almost always favour content that provokes a strong emotional response, because that content holds people’s attention longer. In an election context, that means a structural bias towards outrage and polarisation, not because a designer intended it that way, but simply because it works best for the metrics the system optimises.

Technology has the power to decide who gets to be visible in the public arena.

What is presented today as the prevailing opinion can in reality be the result of finely tuned algorithms amplifying a select few voices while others remain in the shadows. In election campaigns, this means the framing of who counts as a legitimate leader becomes increasingly dependent on technology nobody fully understands. Not truth, but strategy and the power to spread information, then sets the agenda. The danger is that not only political competition comes under pressure, but the basic democratic principles themselves.

This is not a hypothetical scenario. Research into disinformation and personalised political influence documents how microtargeting, tailoring political messages to individual psychological profiles, can present voters with different, sometimes mutually contradictory versions of a candidate, without a shared public debate emerging in which those versions can correct each other. What is missing is not only transparency about who sends a message, but transparency about why this particular person receives this particular message at this particular moment.

Leadership in this landscape calls for a systemic reconsideration of who, as a society, is given room and who is excluded. It calls for awareness of how technology changes the dynamics of power, visibility and influence, not as an abstract problem but as something that recurs every election cycle, faster than the time before.

Imagine you are responsible for a platform that allows political parties to spread personalised messages through automated accounts. The business advantages are clear. But you realise an ethical dilemma lies beneath it: how much influence should technology have over a democratic process, and how do you find the balance between promoting your product and protecting the integrity of the election it affects?

Responsible leadership here calls for the formation of conscience, not a product update.

It calls for transparency about how technology is deployed, and for guidelines that make the ethical limits of influence explicit rather than implicit. It calls for actively thinking through the impact a company can have on the larger system it operates within, not only once damage has occurred, but before the product is launched. That means sparking conversations about the ethics of technology, internally and externally, and ensuring that products serve not only users, but also the wider society those users live in.

What are you sustaining, even with the best of intentions, that further distorts the public space? That is not a question reserved exclusively for executives of technology companies. It is a question for anyone who shares messages, takes part in online discussions, or works within an organisation that uses digital communication tools for influence, even when that influence targets customers rather than voters.

Leadership in the world of AI and digital influence does not primarily call for technical expertise. It calls for moral clarity and the capacity to place choices within a wider societal perspective. Technology without reflection tempts towards manipulation. Technology with conscience can be a force for good. The difference between the two is rarely determined by the technology itself, but by the people who decide how it is used.

Notes for those who wish to read further:

1. Shoshana Zuboff, The Age of Surveillance Capitalism (2019, PublicAffairs). On how systems optimised for attention structurally favour content that provokes emotion and outrage.
2. Cass Sunstein, #Republic: Divided Democracy in the Age of Social Media (2017, Princeton University Press). On how personalised information delivery undermines the shared public debate a democracy needs.
3. Samuel Woolley & Philip Howard, Computational Propaganda: Political Parties, Politicians, and Political Manipulation on Social Media (2018, Oxford University Press). On the workings and global impact of automated political influence.
4. Zeynep Tufekci, Twitter and Tear Gas: The Power and Fragility of Networked Protest (2017, Yale University Press). On how digital platforms redistribute the visibility and invisibility of political voices.
5. Luciano Floridi, The Ethics of Information (2013, Oxford University Press). On the moral infrastructure information systems form, even when that infrastructure remains invisible to the people it influences.