The robots who predict the future
View original at technologyreview.comMIT Technology Review - Ai Research Title: The robots who predict the future Date: 2026-02-18 11:00 Source: https://www.technologyreview.com/2026/02/18/1132579/robots-predict-future-book-review/ <p>To be human is, fundamentally, to be a forecaster…
What we drew from this source
The claims Via News extracted from this document. We point to the source; we don't replace it.
In an optimization-driven reality, every life decision is posed as if it were a round at an imaginary casino, and every argument can be reduced to costs and benefits, means and ends
80% confidencePredictive algorithms represent the system working as intended, not unintended consequences; if an algorithm promotes outrage to maximize engagement and ad clicks, that is because promoting outrage is good for profits from ad sales
80% confidenceThe only way to counter harmful predictive AI is through broad democratic control over the means of prediction: data, computational infrastructure, technical expertise, and energy
80% confidenceThe ubiquitous, mostly invisible predictive layer grafted onto daily life by profit-hungry corporations is leading to a crueler, blander, more instrumentalized world where life's possibilities are foreclosed, age-old prejudices are entrenched
80% confidenceSocieties that rely most heavily on prediction tend toward oppression and authoritarianism
80% confidenceHumans made major scientific and societal advances — clean water, antibiotics, public health improvements, physics breakthroughs, cars, airplanes, modern democracy — without formal systems of rationality or optimal decision theory
80% confidenceMathematical rationality — the narrow statistical conception of optimal decision-making — took firm hold at the tail end of World War II, when mathematical models proved useful against the Axis powers and convinced scientists they could be the basis for designing the first computers
80% confidencePredictions are veiled prescriptive assertions that tell us how to act; they are speech acts, and when we believe a prediction and act in accordance with it, it is akin to obeying an order
80% confidenceWorking to create less biased, more equitable algorithms will not fix the harms of predictive AI because the algorithms rely on past data that is often racist, sexist, and flawed, and profit incentives will always trump attempts to eliminate harm
80% confidenceThe core algorithms developed in the post-WWII period drive the automated decisions of the modern world, whether managing supply chains, scheduling flight times, or placing advertisements on social media feeds
80% confidenceA prediction is a magnet that bends reality toward itself; when the force of the magnet is strong enough, the prediction becomes the cause of its becoming true
80% confidenceTechnology is not destiny; humans make it and choose how to use it, and defying uninvited daily predictions may be the most appropriate human response
80% confidencePredictions can distract us from the challenges of the present; AI promises of artificial general intelligence shift attention away from very real and pressing problems that AI is currently causing
80% confidence