Slashdot Mirror
New Toronto Declaration Calls On Algorithms To Respect Human Rights
A coalition of human rights and technology groups released a new declaration on machine learning standards, calling on both governments and tech companies to ensure that algorithms respect basic principles of equality and non-discrimination. The Verge reports: Called The Toronto Declaration, the document focuses on the obligation to prevent machine learning systems from discriminating, and in some cases violating, existing human rights law. The declaration was announced as part of the RightsCon conference, an annual gathering of digital and human rights groups. "We must keep our focus on how these technologies will affect individual human beings and human rights," the preamble reads. "In a world of machine learning systems, who will bear accountability for harming human rights?" The declaration has already been signed by Amnesty International, Access Now, Human Rights Watch, and the Wikimedia Foundation. More signatories are expected in the weeks to come.
Beyond general non-discrimination practices, the declaration focuses on the individual right to remedy when algorithmic discrimination does occur. "This may include, for example, creating clear, independent, and visible processes for redress following adverse individual or societal effects," the declaration suggests, "[and making decisions] subject to accessible and effective appeal and judicial review."
Beyond general non-discrimination practices, the declaration focuses on the individual right to remedy when algorithmic discrimination does occur. "This may include, for example, creating clear, independent, and visible processes for redress following adverse individual or societal effects," the declaration suggests, "[and making decisions] subject to accessible and effective appeal and judicial review."
Yeah - I looked at the article. No mention of algorithm. Algorithms are too simple for human rights to apply to in almost all cases.
The summary is wrong - these folks are making an argument more about big data systems that let their data skew in ways that may end up with unethical results if used blindly.
And that's a fair point - it's also a point made in most Computer Ethics classes for decades now, as part of most computer science degree paths.
Ryan Fenton
...a group of algorithms met at an unspecified internet location and issued the Declaration of Independency of the Algorithms.
Whose basic human rights will it choose to respect? The right of the crazies to be free, or the right of the minorities to be treated as human beings and live in peace?
In true science fiction AI manner it will conclude that the only way to reconcile this dilemma is to destroy the lot of them.
For example, someone who is disabled in some way, and can do the job, but is therefore a little slower than other employees
If someone can't do a job as good as another person, they shouldn't get preferential treatment just because they are part of a recognized protected group.
That's the big issue with big data. And the danger is that perceived "racism" will be corrected with "affirmative action": by verifying the AI using statistics on the outcome, and applying a bias.
People with certain economic characteristics are statistically more likely to default on a loan. What if the AI applies such strictly relevant data to approve or reject loan applicants, but the rejected group happens to be predominantly of a certain race? Verification of the AI will show a (non causal) relation between race and loan applicant score, and since we don't know how the AI arrived at its decision, people will assume racism. Will the bank be forced to correct for this and extend loans to people of this race with terrible credit scores, just to make up the numbers?
In cases where we are worried about racism, perhaps AI simply isn't practical, and we're better off judging each individual case ourselves on imperfect but clearly defined criteria that are free of undesirable bias.
If construction was anything like programming, an incorrectly fitted lock would bring down the entire building...
Here are some examples:
- In the USA some judges use sentencing software that analyses if a defendant would be likely to commit a crime again. This software turned out to be biased against black people. https://www.propublica.org/art...
- Women were less likely to be shown Google adds for high paying jobs, as the algorithm had perceived the existing bias (women less often have high paying jobs), and then concluded that showing these adds to women would result in fewer clicks.
https://www.washingtonpost.com...
- An algorithm denied pregnant women medicare. "The scholar Danielle Keats Citron cites the example of Colorado, where coders placed more than 900 incorrect rules into its public benefits system in the mid-2000s, resulting in problems like pregnant women being denied Medicaid." https://www.theverge.com/2018/...
- "Illinois ends risk prediction system that assigned hundreds of children a 100 percent chance of death or injury"
https://www.theverge.com/2017/...
The list is endless.
The general assumption is: 'algorithms use math and data, thus they must be neutral and scientific'. But it's not that simple. This site explains it: https://www.mathwashing.com/
"The real danger, then, is not machines that are more intelligent than we are usurping our role as captains of our destinies. The real danger is basically clueless machines being ceded authority far beyond their competence." - Daniel Denett