AI Safety Thursdays: How You Can Help Steer Us Toward a Positive AGI Future — LessWrong