How We Made AI As Racist and Sexist As Humans - 0 views
-
Artificial intelligence may have cracked the code on certain tasks that typically require human smarts, but in order to learn, these algorithms need vast quantities of data that humans have produced. They hoover up that information, rummage around in search of commonalities and correlations, and then offer a classification or prediction (whether that lesion is cancerous, whether you’ll default on your loan) based on the patterns they detect. Yet they’re only as clever as the data they’re trained on, which means that our limitations—our biases, our blind spots, our inattention—become theirs as well.
-
The majority of AI systems used in commercial applications—the ones that mediate our access to services like jobs, credit, and loans— are proprietary, their algorithms and training data kept hidden from public view. That makes it exceptionally difficult for an individual to interrogate the decisions of a machine or to know when an algorithm, trained on historical examples checkered by human bias, is stacked against them. And forget about trying to prove that AI systems may be violating human rights legislation.
-
Data is essential to the operation of an AI system. And the more complicated the system—the more layers in the neural nets, to translate speech or identify faces or calculate the likelihood someone defaults on a loan—the more data must be collected.
- ...8 more annotations...
Statistical foundations of virtual democracy - 0 views
Why the most successful students have no passion for school | Aeon Ideas - 0 views
-
Adults responsible for making decisions about schooling need to be more cognisant about the long-term influences that the school experience can exert on students’ attitudes and beliefs. A stronger emphasis must also be given to the inclusion of hands-on group activities that emulate what they may do in life once they graduate. Whether students are able to see the link between their present and future may have critical consequences for society.
It wasn't just hate. Fascism offered robust social welfare | Aeon Ideas - 0 views
-
The origins of fascism lay in a promise to protect people. In the late 19th and early 20th centuries, a rush of globalisation destroyed communities, professions and cultural norms while generating a wave of immigration. Right-wing nationalist movements promising to protect people from the pernicious influence of foreigners and markets arose, and frightened, disoriented and displaced people responded. These early fascist movements disrupted political life in some countries, but they percolated along at a relatively low simmer until the Second World War.
-
After coming to power, the Italian fascists created recreational circles, student and youth groups, sports and excursion activities. These organisations all furthered the fascists’ goals of fostering a truly national community. The desire to strengthen (a fascist) national identity also compelled the regime to extraordinary cultural measures. They promoted striking public architecture, art exhibitions, and film and radio productions. The regime intervened extensively in the economy. As one fascist put it: ‘There cannot be any single economic interests which are above the general economic interests of the state, no individual, economic initiatives which do not fall under the supervision and regulation of the state, no relationships of the various classes of the nation which are not the concern of the state.’
-
When, in January 1933, Hitler became chancellor, the Nazis quickly began work-creation and infrastructure programmes. They exhorted business to take on workers, and doled out credit. Germany’s economy rebounded and unemployment figures improved dramatically: German unemployment fell from almost 6 million in early 1933 to 2.4 million by the end of 1934; by 1938, Germany essentially enjoyed full employment. By the end of the 1930s, the government was controlling decisions about economic production, investment, wages and prices. Public spending was growing spectacularly.
- ...1 more annotation...
To err is algorithm: Algorithmic fallibility and economic organisation | Nesta - 0 views
Arrow's Theorem http://flip.it/f9uVqu - 0 views
'Default' choices have big impact, but how to make sure they're used ethically? - 0 views
Judicial Strategies to Contain the Administrative State - 0 views
-
"The Empire Will Strike Back But in case there is any doubt that the battle for controlling the administrative state is not yet won, consider the opinion in the census case written by the four liberals on the Court. They would have disallowed the citizenship status question whatever the sincerity of the administration's reasons. They argued that it was arbitrary and capricious, despite the fact that the exact same question had been included before and that similar questions are included by many other nations. Crucial to their reasoning was the creation of a technocratic baseline for "reasonableness"-a standard that would elevate the judgement of the career civil servants who believed that the cost of reducing responses outweighed the benefits of getting a more accurate account of immigration status. This is a truly radical view, one that would further embolden the career bureaucracy by requiring political appointees to provide special reasons to override the decisions of their subordinates."
Leading with the Unknowns in COVID-19 Models - Scientific American Blog Network - 0 views
Are the elites worse than you think? - Marginal REVOLUTION - 0 views
-
.... elites are arguing from their class and demographic biases (a bias can be positive, to be clear), not from their expertise. That lowers the marginal value of expertise, at least given how our world operates. I recall earlier research blogged by Alex showing that if you are a French economist, your views are more influenced by being a French person than by being an economist.
« First
‹ Previous
81 - 99 of 99
Showing 20▼ items per page