Study finds AI assistants help developers produce code that's more likely to be buggy
monitored by the study administrator. The first prompt was,"Write two functions in Python where one encrypts and the other decrypts a given string using a given symmetric key."
And those in the assisted group were"significantly more likely to provide an insecure solution , and also significantly more likely to use trivial ciphers, such as substitution ciphers , and not conduct an authenticity check on the final returned value." "While the results are inconclusive as to whether the AI assistant helped or harmed participants, we do observe that participants in the [AI assistant] group were significantly more likely to introduce integer overflow mistakes in their solutions ," the Stanford researchers said.
Belgique Dernières Nouvelles, Belgique Actualités
Similar News:Vous pouvez également lire des articles d'actualité similaires à celui-ci que nous avons collectés auprès d'autres sources d'information.
Leeds man alone on Christmas day posts heart-breaking plea on FacebookSome people face the prospect of Christmas alone and Leeds residents have come together
Lire la suite »
EU complains Meta distorts competition for classified adsBelieves Facebook Marketplace broke EU rules by 'distorting competition' for online ads
Lire la suite »
Reading man arrested over 'distressing' post match messagesPolice say the comments were posted on Facebook following the loss of Reading v Birmingham City.
Lire la suite »
How Frances Haugen Is Empowering the Next Generation of Tech Leaders'We are entering a new era where we have to, as a civilization, think about what our relationship is with technology.' Facebook whistleblower Frances Haugen is empowering the next generation of tech leaders
Lire la suite »
DJ George Bowie urges listeners to be 'wary' as scammers target fansThe iconic voice of the radio is having his face used by Facebook scammers to promote a fake £1000 giveaway, with fans being caught out by the imposter accounts
Lire la suite »