Task 11: Detain/Release

I decided to do Option 1, the Detain/Release simulation, for Task 11.

At the beginning of the simulation, I spent a lot more time carefully reading the charges for each defendant and their statements, but later on relied a lot more on the low/medium/high risk assessment when deciding whether or not to detain or release the defendant. Towards the end of the simulation, I found myself almost mindlessly selecting “detain” when I saw yellow for medium and red for high when it came to the defendants’ risk assessment in terms of committing a crime and violence. When defendants I released appeared in the newspaper headlines, I became more cautious again with my rulings. I also felt some pressure to release more defendants and be more lax with my rulings when I noticed the jail capacity bar filling up.

While I avoided looking at the blurred faces, the ages, and the sex of the defendants to ensure that I would not be biased in any way when making my decisions, I ended up being more careful with my rulings for defendants that had statements concerning family compared to those that had statements about housing or work; however, for defendants that committing assault or more violent crimes, I detained them regardless of their personal statements.

The most terrifying thing I noticed from doing the Detain/Release simulation is the fact that it almost felt like a game to me, where I felt pressured to keep jail capacity and fear low and at times relied too much on the risk assessment. As Porcaro (2019) stated, “the mere presence of a risk assessment tool can reframe a judge’s decision-making process”, which is definitely what happened when I was doing this simulation. I am also terrified by the thought of real-world judges in pretrial hearings potentially being influenced by these AI risk assessments without realizing they are. Are we relying too much on technology, to an extent where our humanity is being compromised?

References