A technological promise can be true and still be presented to serve particular interests. Reducing the time needed for a task says nothing about who benefits from the saving, which work moves to other people or how much correcting errors will cost. Evaluating a product requires examining both its performance and the economic arrangement around it. Enthusiasm tends to display the saving more clearly than the complete bill.
Emily M. Bender and Alex Hanna direct The AI Con against commercial exaggeration and the interests it can conceal. The publisher's presentation invites scrutiny of harms, the distribution of benefits and corporate power, alongside capability claims. It calls for contesting adoption conditions. Source: The Bodley Head · Penguin Random House UK
A saving can change pockets
Consider a hypothetical insurer advertising much faster service. Its system answers immediately but repeatedly requires complainants to reconstruct their cases. The company has reduced customer-service hours; customers have increased their own working hours. An internal improvement can coexist with a worse service. Finding out what happened requires measuring whether the problem was actually resolved, rather than merely the speed of the first reply.
Those differences require separating private costs from consequences falling on third parties. Part of the problem may lie outside the agreement between a tool's purchaser and its seller. If affected people do not participate in the transaction, their time and difficulties may carry little weight in the decision. The contract's apparent efficiency can depend partly on the burdens it manages not to recognise.
A strong criticism should follow that mechanism to the evidence. Do resolution times fall? Who abandons the procedure? Does the comparison include difficult cases? Is an accessible alternative available? The answers may reveal a real but unevenly distributed improvement, a broken promise or a combination. Suspecting the seller's interests helps formulate questions; it does not replace their answers. Benefiting from a technology does not, on its own, establish deception.
Leaving has a price too
Commercial power also appears when customers can no longer leave easily. Migration may require recovering data, rebuilding procedures and teaching workers another system. An initially cheap offer can become expensive even if its nominal price stays unchanged. Anyone evaluating the purchase needs to know the exit conditions before dependency turns an available alternative into an impracticable one. The relevant comparison includes the customer's future ability to choose.
Regulatory responses deserve the same care. Requiring intelligible documentation or comparable testing can improve competition. Imposing procedures only large firms can finance could strengthen the actors the rules are intended to constrain. Criticism of technological power becomes more precise when it distinguishes protecting people from selecting a procedure that inadvertently reinforces the dominant supplier. The measure's name provides no guarantee of its effect on the market or on those needing protection.
The main question this recommendation brings to readers is not whether they should be excited or outraged by a demonstration. It is which concrete claim they can check, against what comparison and whose lives bear the consequences of getting it wrong. Useful technology can be sold through exaggeration; a persuasive denunciation can need better evidence. Keeping both possibilities open allows debate over the purchase, the contract and the regulation without handing the conclusion to the sales campaign or its opposing slogan.