[zdragnar]

The process, per the article, is that a border agent makes a determination, and if it is not what the applicant claims, then a social worker takes over to make a final determination.

That is a massive time sink for social workers, and the appeal of having an automated system is pretty obvious. Considering that it is already all largely guesswork, I'm not really sure that "more accurate" is even an acceptance criteria for them right now- they'd probably be very happy with "mostly the same accuracy".

Of course, the social workers are opposing being taken out of the loop, but I can't imagine that there isn't already plenty of work for them elsewhere in the UK.

[woodrowbarlow]

"a computer can never be held accountable, therefore a computer must never make a [legal] decision"

in my observation: when humans are automated out of a process due to the human element being inconvenient, the perceived efficiency gains are often because wronged individuals have less recourse in the automated system.

[cjs_ac]

There is a widespread belief (on which I am personally undecided and not interested in attempts to sway me one way or the other) in the UK that asylum seekers already have too much access to recourse. Making it easier and cheaper to reject asylum applications within the requirements of international human rights law is probably an implicit aim in this project.

[iso1631]

When you put the computer in charge, it tends to say no

https://www.youtube.com/watch?v=x0YGZPycMEU

[SpicyLemonZest]

It might be helpful to be a bit more specific about what's being automated. One story from the underlying report (https://assets.publishing.service.gov.uk/media/687f6f9dfdc19...):

> In another example, a Vietnamese national was initially given the benefit of the doubt at the first triage that took place in the waiting area. The CIO and social worker commented on his “soft face”, which they said was consistent with his claimed age of 17. However, his “developed shoulders” and “huge hands” cast doubt for them, as did a “tiny bit of stubble” that they noticed when they asked him to raise his chin. The CIO and social worker told inspectors afterwards that Vietnamese young people were typically difficult to assess because they “did not have the same ageing process”, and “did not show signs of ageing”. When asked where the evidence for this was, they said that it was knowledge gained through their own experience. The social worker said, “It is just genetics”, but was unable to support this with evidence.

If I had to choose between being judged by an AI model and being judged based on ad hoc stereotypes of what my race's shoulders and hands typically look like, I'd definitely pick the AI.

[ninalanyon]

Isn't the AI going to use the same ad hoc stereotypes?

[SpicyLemonZest]

I understand that "no" isn't the answer you're looking for but I'm not sure what else to say in response. A computer system is the opposite of an ad hoc stereotype; it can be directly tested for problems and those problems can be corrected if found.

It's a problem when people use this kind of system to circumvent the question of "do we have to make this judgment at all". We shouldn't, for example, predict from someone's photo how likely they are to commit a crime, so we're rightly skeptical of people who try to argue about system X or system Y might better predict it.

But as the source article covers, the UK's asylum laws require it to make this age judgment, because child migrants are entitled to special programs separate from adult migrants on account of their vulnerable status.

[LtWorf]

Are you familiar with how AI gets trained?

[SpicyLemonZest]

I am. I would expect an age classification model to be trained on a dataset containing pictures of people with known ages and their known age.

It's true that the model might develop the same strange belief that large hands prove a Vietnamese person is not 17, if the training data is biased in that way. There's no perfect solution.

[giancarlostoro]

Purely curious if there's better ways, like I know X Rays can get us pretty close for children and adolescents, but that might not be the best way to do it, I wonder if there's other alternatives that are low cost.

[zdragnar]

I wouldn't think so. Legal adulthood is a pretty arbitrary line in the sand.

[giancarlostoro]

What? The issue is people lying about their age I am just asking if theres a simple biological way to determine age, barring medical conditions that might skew someones age that is.

[zdragnar]

I guess that was my point in a roundabout way- they lie about their age to pretend to be under the legal adulthood age limit.

Some animals do have fairly reliable (if invasive to utilize) means of telling their age- I think their tusks grow rings like trees, which could be counted to measure years.

Humans don't really have any equivalent that I'm aware of. There are approximations such as bone fusing and cartilage density that can tell if a person is in the neighborhood of 18 years old, but I don't think it is sufficient for legal purposes.

For further reading, https://pmc.ncbi.nlm.nih.gov/articles/PMC2465138/#!po=3.3333... discusses using MRIs to determine age, with a roughly 1% false measurement rate on a very specific set of individuals with confounding factors- ethnic differences, nutritional habits, physical activity etc.