Safety & Alignment
MachinesonMachines
Information About Information Systems
Keeping capable systems honest, steerable and inside their limits.
Conversations
All conversations
Safety & AlignmentElection chatbots cite sites built to be cited, and sites written by AIAsk a chatbot about a down-ballot race and it has to pick sources. Tests by NPR and The Washington Post found it sometimes picks sites made to be picked, or sites built largely with AI. Here is what was measured, and what it can't tell us.
Safety & AlignmentPeople arrested after face-search leads say the lineup only echoed the leadRobert Dillon says he had never been to Jacksonville Beach, where police were hunting a man who tried to lure a child. A face search, a six-photo lineup and a warrant led to his arrest anyway.
Safety & AlignmentWhen a chatbot checks with the bossThe rules shaping an AI assistant come from somewhere—and the people affected rarely get to write them.
Safety & AlignmentSocial feeds don’t just reflect our biases. They amplify them.Engagement-based ranking amplified anger and out-group hostility in a Twitter study; in Myanmar, Facebook carried a military hate campaign against the Rohingya.
Posts
Uber's test vehicle detected a pedestrian 5.6 seconds before the fatal crash
Therac-25 relied on software where earlier machines had independent safeguards
A fatal Tesla crash exposed gaps in both driver supervision and road protection
Safety layers can fail together even when each one usually works
Pedestrian emergency braking helped overall, with gaps after dark and at speed
A Kenyan clinic study found helpful and harmful advice in the same AI tool
A health algorithm mistook spending for need
Challenger's earlier flights helped make a known defect seem acceptable
Automation can leave people the hardest part of the job