What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
AI does not have a political ideology or intent. The danger Kate Crawford raised in a 2017 SXSW talk was that people in power could use data systems to track, classify and target populations while presenting the systems’ decisions as neutral and beyond challenge.
What was Kate Crawford warning about?
In March 2017, researcher Kate Crawford gave a SXSW session titled “Dark Days: AI and the Rise of Fascism.” The talk asked how AI and machine learning might be used in a period of rising authoritarian politics, and how society could protect people most at risk. SXSW’s recording is the primary source for the session.
As an Amazon Associate I earn from qualifying purchases.
In a contemporaneous report, The Guardian described Crawford’s concern that authoritarian actors could use AI to centralize power and target populations. The phrase “a fascist’s dream” is a warning about that political use of technology—not a claim that AI itself has fascist beliefs or intentions. The Guardian’s report quoted Crawford describing the political context: “Just as we are seeing a step function increase in the spread of AI, something else is happening: the rise of ultra-nationalism, rightwing authoritarianism and fascism.”
How can data systems concentrate power?
Large-scale data systems can make it easier for institutions to sort people into categories, monitor them, and act on those classifications. When the people being classified cannot see how a system works or challenge its decisions, the technology can reinforce power imbalances. Crawford’s concern was about the institutions deploying these tools and the consequences for the people subjected to them.
#1 Best Overall
- Tracking: Data collection and automated analysis can help institutions follow people or groups at scale.
- Classification: Systems can assign categories or risk labels that shape how people are treated.
- Targeting: Those classifications can be used to direct scrutiny or other actions toward particular populations.
- Unaccountable authority: Describing a system as objective can obscure the human choices behind its data, design and use.
The AI Now Institute’s 2017 report offers related historical context on identification and population-documentation systems, including NSEERS and the Book of Life. It is context for the risks of classification and documentation, not a current measurement of AI deployment.
Why “neutral” AI deserves scrutiny
Machine-learning systems learn patterns from data. When that data reflects human judgments or unequal treatment, a model can carry those patterns into its outputs. Automation does not, by itself, remove the assumptions embedded in the data or in decisions about how a system is built and used.
Rank #2
As quoted by The Guardian, Crawford put it this way: “We should always be suspicious when machine learning systems are described as free from bias if it’s been trained on human-generated data.” Calling a system neutral is not evidence that it is fair; scrutiny requires asking what data and choices shaped it, who is affected, and who can challenge its decisions.
What the facial-analysis example does—and does not—show
Futurism’s 2017 coverage discussed a facial-analysis example raised in connection with Crawford’s talk. That example should be understood as part of the historical warning, not as proof that facial appearance can reliably predict criminality. The reporting available here does not establish the original study’s methodology or validity, so it cannot support that conclusion.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.What the 2017 warning establishes today
Crawford’s talk and the reporting around it establish a warning about how AI and large-scale data systems could serve political aims: tracking people, classifying them, targeting groups and concentrating power. They do not establish how widespread or effective particular AI surveillance systems are today, and they provide no prevalence estimate for AI-enabled authoritarian abuse. The evidence is historical and should not be mistaken for a current deployment survey.
The practical test for any such system is therefore not whether it is called AI or described as neutral, but whether its use is transparent and accountable, whether affected people can challenge decisions, and whether safeguards prevent population-level targeting.
Quick Recap
Rank #4
Sources and further context
- Futurism’s March 14, 2017 report on Crawford’s warning.
- The Guardian’s March 13, 2017 report, updated January 6, 2021, including the quoted remarks.
- SXSW’s recording of “Kate Crawford: DARK DAYS: AI and the Rise of Fascism – SXSW 2017,” published June 7, 2017.
- AI Now Institute’s 2017 Report for related historical context.
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
Quick wins for a faster PC:
Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Clear out junk files and repair common Windows errorsFree Scan →Scan for outdated or missing drivers - takes under a minuteDriver Scan →
