Atlas of AI, book review: Mapping out the total cost of artificial intelligence

Atlas of AI: Ability, Politics, and the Planetary Prices of Artificial Intelligence • By Kate Crawford • Yale University Push • 336 web pages • ISBN: 978–three hundred-20957- • £20   

“Question forgiveness, not permission” has extended been a guiding theory in Silicon Valley. There is no technological subject in which this theory has been a lot more practiced than the machine understanding in modern day AI, which is dependent for its existence on giant databases, pretty much all of which are scraped, copied, borrowed, begged, or stolen from the giant piles of facts we all emit day-to-day, knowingly or not. But this facts is rarely at any time rigorously sourced with the subjects’ permission.  

“Because we can,” two sociologists inform Kate Crawford in Atlas of AI: Ability, Politics, and the Planetary Prices of Artificial Intelligence, by way of acknowledging that their academic institutions are no distinctive from know-how firms or governing administration organizations in regarding any facts they obtain as theirs for the taking to prepare and check algorithms. Photographs grow to be infrastructure. This is how machine understanding is created. 

All people would like to communicate about what AI is good or hazardous for — determining facial images, interpreting speech instructions, driving cars and trucks (not however!). Many want to pour ethics above modern AI, as if making procedures could alter the navy funding that has outlined its essential mother nature. Few want to focus on AI’s real prices. Kate Crawford, a senior researcher at Microsoft and a research professor at the University of Southern California, is the exception. 

SEE: Making the bionic mind (totally free PDF) (TechRepublic)

In Atlas of AI, Crawford begins by deconstructing the renowned competition that ‘data is the new oil’. Typically, that prospects persons to communicate about data’s financial worth, but Crawford focuses on the fact that both equally are extractive systems. Extraction is mining (as in ‘data mining’ or oil wells), and in which mining goes, so observe environmental destruction, human exploitation, and profound society-vast effects.  

Crawford underlines this stage by heading to Silver Peak, Nevada, to check out the only running lithium mine in the US. Lithium is, of system, a vital ingredient in battery packs for anything from smartphones to Teslas. Crawford follows this up by thinking about the widening implications of extraction for labour, the resources of facts, classification algorithms, and the nation-state behaviour it all underpins, ending up with the electric power structures enabled by AI-as-we-know-it. This way lies Project Maven and ‘signature strikes’ in which, as previous CIA and NSA director Michael Hayden admitted, metadata kills persons. 

Snake oil

But some of this is patently false. Crawford traces again the image datasets on which the most current disturbing snake oil — emotion recognition — is based mostly, and finds they were developed from posed pics in which the subjects were instructed to provide exaggerated illustrations of emotional reactions. In this circumstance, ‘AI’ is made all the way down. Is there, as Tarleton Gillespie questioned about Twitter trends, any actual human reflection there? 

Even though other know-how books have tackled some of Crawford’s subjects (as well several of which have been reviewed listed here to listing), the closest to her integrated structural solution is The Prices of Link by Nicholas Couldry and Ulises A. Mejias, which sights our current technological reconfiguration as the beginnings of a new partnership concerning colonialism and capitalism. 

“Any sufficiently advanced know-how is indistinguishable from magic,” Arthur C. Clarke famously wrote. Subsequent Crawford, this looks a lot more like: “Any know-how that looks like magic is hiding one thing.” So several darkish strategies lie in how the sausage is created. 

New AND Related Content

The Prices of Link, e book evaluation: A broader perspective of surveillance capitalism

The greatest expense in database historical past, the greatest social community at any time, and other graph tales from Neo4j

Now Google is employing AI to design chips, much more rapidly than human engineers can do the position

Drones with exoskeletons experience off in soccer obstacle

Terrified, human? Emotion-detection AI meets eye-monitoring know-how

Read through a lot more e book assessments