By Crispin Kaheru, Member of the Uganda Human Rights Commission (UHRC)
Whenever people speak of a stolen election, what often comes to mind are stuffed ballot boxes and forged tally sheets. Today, world over, it may look a little different. The concept is the same though, but the method has changed. Elections are now stolen quietly and invisibly, through lines of code and remote servers sometimes mounted thousands of kilometres away.
In the digital age, the most vulnerable polling station is no longer a rural school or a crowded playground in a trading centre, it is inside a network, inside the internet, in the cyber-space. When that network comes under a verified external cyber-attack, the state faces a grave choice, either to act decisively to protect the vote, or risk presiding over the silent theft of the people’s will.
This is not a hypothetical concern. Across the world, elections have become prime targets for interference, cyber-espionage, and coordinated disinformation campaigns. Voter registers are being probed, results-transmission systems are being scanned, and digital reporting platforms are being flooded with false data designed to confuse, discredit, or delegitimise outcomes. Such attacks are not acts of free expression. They are assaults on the most fundamental political right, the right of citizens to choose their leaders through genuine elections.
Uganda’s general election of 15 January 2026 will not take place in an analogue vacuum. It is already unfolding in a hyper-connected environment where results transmission, tallying, verification, observation, media reporting, and voter mobilisation rely heavily on digital infrastructure. If that infrastructure is deliberately compromised by (external) actors, the harm is immediate and profound. A hacked election is not merely an administrative failure, it is a human rights violation that robs citizens of political agency and risks triggering unrest and instability.
At the same time, the internet itself has become a rights space. Freedom of expression, access to information, media freedom, and civic participation increasingly depend on connectivity. Internet shutdowns have therefore attracted global criticism, especially where they punish entire populations for the actions of a few unscrupulous actors.
But human rights law has never required states to be passive in the face of imminent harm. As I have often argued in this column, rights exist in balance, not isolation. The right to expression does not include the right to sabotage an election. The right to access information does not extend to malicious interference with sovereign democratic processes. When credible and verifiable evidence shows that hackers are actively attempting to compromise election systems, the state acquires not just the power, but the duty, to protect the integrity of the vote.
This is where the debate must mature. The question is not whether internet restrictions are good or bad in the abstract. The real question is whether, in exceptional circumstances, a temporary interruption can prevent a far greater rights catastrophe, which in this case is a stolen election whose consequences may linger for years. Around the world, governments have imposed temporary internet shutdowns during moments of acute national risk. Uganda itself has previously confronted this difficult tension.
Any consideration of switching off or limiting internet access must therefore be triggered by verified technical assessments showing an imminent and serious cyber threat to election-critical systems. The purpose must be singular, to protect the vote, not to manage politics.
Even then, the response must be proportionate. The first duty of the state is to exhaust all less-restrictive measures, such as blocking malicious traffic and neutralising specific attack vectors. Only where such targeted defences fail or are overwhelmed should a shutdown be considered. And where it is used, it must be clearly communicated, and subject to legal oversight. Citizens deserve to know why such a step has been taken.
Doing nothing in the face of cyber-interference is not neutrality; it is negligence. Elections are a human-rights event, and protecting them may sometimes require uncomfortable, temporary measures.
So, in the event that cyber-interference poses a real and immediate threat to the January 2026 elections, a carefully calibrated internet shutdown may be justified in the public interest. But it must be a shield, not a hammer; a pause to protect the vote, not a silence imposed to control it.
Ultimately, the state does not earn legitimacy by pretending risks do not exist. It earns legitimacy by confronting those risks openly, and lawfully. Protect the internet, yes. But when the ballot itself is under attack, protect the vote first and do so in a way that keeps human rights intact when the lights come back on.
