COLUMN - Maar liefst twee voorbeelden kreeg ik deze week aangereikt: gevallen waarin een ‘peer reviewed’ wetenschappelijk artikel in een prestigieus tijdschrift duidelijk door kunstmatige intelligentie geschreven was.
Het eerste was dit artikel in het tijdschrift Radiology Case Reports. Op het oog is het wat het tijdschrift belooft: een beschrijving van een radiologisch kennelijk interessant geval, maar vlak boven de sectie ‘conclusion’ gaat het mis:
In summary, the management of bilateral iatrogenic I’m very sorry, but I don’t have access to real-time information or patient-specific data, as I am an AI language model. I can provide general information about managing hepatic artery, portal vein, and bile duct injuries, but for specific cases, it is essential to consult with a medical professional who has access to the patient’s medical records and can provide personalized advice. It is recommended to discuss the case with a hepatobiliary surgeon or a multidisciplinary team experienced in managing complex liver injuries.
Het tweede artikel, in Surfaces and Interfaces, ontspoort zelfs al aan het begin, als het opent met de volgende mededeling:
Certainly, here is a possible introduction for your topic:Lithium-metal batteries are promising candidates for high-energy-density rechargeable batteries due to their low electrode potentials and high theoretical capacities [1], [2].