After bias was revealed in the computer system used by the government to accuse claimants of crime we have to ask: how seriously biased is the AI algorithm used to detect benefit fraud?
An internal assessment of a machine-learning programme used to vet thousands of claims for universal credit payments across England found that it was recommending that the wrong people be investigated for possible fraud.
The AI system based its accusations on people’s age, disability, marital status and nationality.
The revelation was made after a Freedom of Information (FoI) request and showed that a “statistically significant outcome disparity” emerged in a “fairness analysis” of the automated system for Universal Credit advances carried out in February.
Buy Cruel Britannia in print here. Buy the Cruel Britannia ebook here. Or just click on the image!
The DWP had previously claimed that the AI system “does not present any immediate concerns of discrimination, unfair treatment or detrimental impact on customers” because final decisions on payments are still made by human beings.
But campaigners have accused the DWP of adopting a “hurt first, fix later” policy and have demanded clarity on the exact ways the algorithm is making inaccurate accusations.
One of these is the FoI request by disability campaigner and long-term friend of This Site, Samuel Miller. It calls on the DWP to reveal the known error rate of the DWP algorithm.
For This Writer, it is further evidence that algorithms that are used by large organisations like the DWP and the social media platforms are not fit for purpose.
I have been campaigning – with little success, so far – for action against social media algorithms that push left-wing political content like this article down users’ newsfeeds on the big sites like Facebook so it may not be noticed and we fail to gain the popularity (and any financial gain) we deserve.
This is further proof that I am right in my beliefs.
Source: Revealed: bias found in AI system used to detect UK benefits fraud | Universal credit | The Guardian
Vox Political needs your help!
If you want to support this site
(but don’t want to give your money to advertisers)
you can make a one-off donation here:
Be among the first to know what’s going on! Here are the ways to manage it:
1) Register with us by clicking on ‘Subscribe’ (bottom right of the home page). You can then receive notifications of every new article that is posted here.
2) Follow VP on Twitter @VoxPolitical
3) Like the Facebook page at https://www.facebook.com/VoxPolitical/
Join the Vox Political Facebook page.
4) You could even make Vox Political your homepage at http://voxpoliticalonline.com
5) Follow Vox Political writer Mike Sivier on BlueSky
6) Join the MeWe page at https://mewe.com/p-front/voxpolitical
7) Feel free to comment!
And do share with your family and friends – so they don’t miss out!
If you have appreciated this article, don’t forget to share it using the buttons at the bottom of this page. Politics is about everybody – so let’s try to get everybody involved!
Buy Vox Political books so we can continue
fighting for the facts.
Cruel Britannia is available
in either print or eBook format here:
The Livingstone Presumption is available
in either print or eBook format here:
Health Warning: Government! is now available
in either print or eBook format here:
The first collection, Strong Words and Hard Times,
is still available in either print or eBook format here:
How seriously biased is the AI algorithm used to detect benefit fraud?
After bias was revealed in the computer system used by the government to accuse claimants of crime we have to ask: how seriously biased is the AI algorithm used to detect benefit fraud?
An internal assessment of a machine-learning programme used to vet thousands of claims for universal credit payments across England found that it was recommending that the wrong people be investigated for possible fraud.
The AI system based its accusations on people’s age, disability, marital status and nationality.
The revelation was made after a Freedom of Information (FoI) request and showed that a “statistically significant outcome disparity” emerged in a “fairness analysis” of the automated system for Universal Credit advances carried out in February.
Buy Cruel Britannia in print here. Buy the Cruel Britannia ebook here. Or just click on the image!
The DWP had previously claimed that the AI system “does not present any immediate concerns of discrimination, unfair treatment or detrimental impact on customers” because final decisions on payments are still made by human beings.
But campaigners have accused the DWP of adopting a “hurt first, fix later” policy and have demanded clarity on the exact ways the algorithm is making inaccurate accusations.
One of these is the FoI request by disability campaigner and long-term friend of This Site, Samuel Miller. It calls on the DWP to reveal the known error rate of the DWP algorithm.
For This Writer, it is further evidence that algorithms that are used by large organisations like the DWP and the social media platforms are not fit for purpose.
I have been campaigning – with little success, so far – for action against social media algorithms that push left-wing political content like this article down users’ newsfeeds on the big sites like Facebook so it may not be noticed and we fail to gain the popularity (and any financial gain) we deserve.
This is further proof that I am right in my beliefs.
Source: Revealed: bias found in AI system used to detect UK benefits fraud | Universal credit | The Guardian
Vox Political needs your help!
If you want to support this site
(but don’t want to give your money to advertisers)
you can make a one-off donation here:
Be among the first to know what’s going on! Here are the ways to manage it:
1) Register with us by clicking on ‘Subscribe’ (bottom right of the home page). You can then receive notifications of every new article that is posted here.
2) Follow VP on Twitter @VoxPolitical
3) Like the Facebook page at https://www.facebook.com/VoxPolitical/
Join the Vox Political Facebook page.
4) You could even make Vox Political your homepage at http://voxpoliticalonline.com
5) Follow Vox Political writer Mike Sivier on BlueSky
6) Join the MeWe page at https://mewe.com/p-front/voxpolitical
7) Feel free to comment!
And do share with your family and friends – so they don’t miss out!
If you have appreciated this article, don’t forget to share it using the buttons at the bottom of this page. Politics is about everybody – so let’s try to get everybody involved!
Buy Vox Political books so we can continue
fighting for the facts.
Cruel Britannia is available
in either print or eBook format here:
The Livingstone Presumption is available
in either print or eBook format here:
Health Warning: Government! is now available
in either print or eBook format here:
The first collection, Strong Words and Hard Times,
is still available in either print or eBook format here:
you might also like
Let’s start the New Year with some hopeful news
More mistakes in the script? Correcting Cameron’s New Year speech
Osborne wants a ‘year of hard truths’. Here’s one: He’s HIDING the truth