The robot predicts exactly the judge's judgment

The UK's successful artificial intelligence research team (AI) can predict a 79% accuracy rate by the European Court of Human Rights.

The research team from three British universities designed a computer algorithm that could predict the ECtHR's accuracy at 79%, IFL Science led the report. Peerj Computer Science said on Oct. 24.

To create this artificial intelligence, the team trained a computer that identifies a common pattern from 584 case files dealt with under Articles 3, 6 and 8 of the European Convention on Human Rights, which mentions the right to be free from torture, fair trial and privacy rights respectively.

AI can accurately predict 79% of the rulings of the European Court of Human Rights. Photo: Posteriori.

By contrasting between the models in this case and the outcome, the machine can then accurately predict court decisions in almost 80 percent of cases.

The algorithm yields the most reliable results when analyzing the "Situation" subsection, which contains details related to the actions and events surrounding a particular case. In contrast, the subsection "Law" refers to the legal arguments put forward by the two parties, the part that gives the lowest accurate prediction.

From there, the team concluded that ECtHR judges were more realistic than formalism, meaning that their decisions were based on facts rather than the legal framework.

With the success of this study, co-author Nikolaos Aletras said that robots with artificial intelligence could be an important tool in identifying cases that are more likely to violate the European Convention on Human rights. This technology can help the court to try more effectively by determining the probability of winning or losing cases, even before they go to court.

"Artificial intelligence will not replace judges and lawyers, but they may find it useful to identify stereotypes in certain cases , " Aletras explained.

AI is especially useful for ECtHR complaints because it gives lawyers and judges a relative view of the probability of success of these appeals. Lawmakers can then make better decisions about whether or not to try a case to save time.