Elliott BledsoeAnthropic have released results open-ended interviews with Claude users about what they want from AI and what they are concerned about. While hopes seem more variable globally, unreliability is almost universally the top concern. It's interesting reading an a great interactive way to share it. Worth a look.
