Brazil 2022 Presidential Election
October 2, 2022 – October 30, 2022
Luiz Inácio Lula da Silva defeated incumbent Jair Bolsonaro by 50.90% to 49.10% in Brazil's closest presidential runoff of the democratic era. The election took place after sustained attacks on electronic voting. Those attacks are documented: the TSE later found Bolsonaro abused political power and misused media in an official event aimed at discrediting the system. The count-manipulation claims are a separate question. TCU comparisons of physical ballot bulletins, published records and integrity tests found no divergence, while the TSE rejected the Liberal Party's post-election challenge for lack of supporting indications.
Official results
First round, 2 October
- Luiz Inácio Lula da Silva48.4%57,259,504
- Jair Bolsonaro43.2%51,072,345
Runoff, 30 October
- Luiz Inácio Lula da Silva50.9%60,345,999
Elected
- Jair Bolsonaro49.1%58,206,354
The Integrity Ledger
What the evidence establishes, and what it does not — side by side, each finding sourced and tiered.
What is established
- Information operationsConfirmed
The incumbent used an official event and public communication channels to attack the electoral system without evidence.
More detail
In a later election-law judgment, the TSE found Jair Bolsonaro responsible for abuse of political power and misuse of media after a meeting with ambassadors that deployed public resources to disseminate unsupported claims about electronic voting. The finding concerned the campaign environment and institutional trust; it was not a finding that votes had been altered.
Sources:1 - Process violationsCorroborated
The election unfolded amid polarization, disinformation, attacks on electoral institutions and political violence.
More detail
The OAS mission documented these pressures while also finding that the TSE and regional electoral courts demonstrated professionalism and institutional robustness in administering the vote.
Sources:1
What is not established
- Result falsificationDebunked
Direct checks found no evidence that the audited electronic records or ballot bulletins had been altered.
More detail
The TCU found no divergence when sampled physical ballot bulletins were compared with published electronic records, no divergence in 641 integrity-test machines and no indication of fraud in the procedures it reviewed. These checks address recorded totals directly; they are stronger evidence than viral screenshots or unexplained numerical patterns.
Sources:1 - Statistical anomaliesDebunked
Different candidate shares across voting-machine models did not establish that older machines were unidentifiable or fraudulent.
More detail
The Liberal Party sought extraordinary verification of machines built before 2020. The TSE found no supporting indication, noted that machines of all models were individually traceable and had been used in both rounds, and confirmed dismissal of the request. A correlation between machine model and candidate share can reflect where different models were deployed; it is not a causal test of manipulation.
Process assessment
Brazil's election administration completed and published the count quickly under intense political pressure. The institutional record is mixed in a precise way: the campaign included a judicially established abuse aimed at discrediting the system, but independent technical comparisons and judicial review did not substantiate claims that the machine count was false.