A new study, shared exclusively with TIME ahead of publication, reveals a promising technique for discouraging AI models from cheating by changing how they're trained.
Follow this section to personalize your feed and get instant alerts. Update your preferences in Account Settings Follow this tag to personalize your feed and get instant alerts. Follow this author to personalize your feed and get instant alerts.
Extract — continue reading at the source.
Related stories
American national sentenced to life in prison for murders of partner and infant daughter in Rome
Myanmar: Civilians under fire from military junta
Harmeet Dhillon, controversial head of DoJ’s civil rights division, steps down
Heavy rains bring flash flooding to Georgia-Tennessee border area
British comedian Russell Brand found guilty of rape and three sexual assaults by London jury
Duchess of Sussex’s estranged half-brother ‘person of interest’ in child sexual abuse inquiry