Yeah I was thinking of the approach used in the invention described in the OP, with a pre-chilled slug used to set up a temporary gradient. In that case you cou

Yeah I was thinking of the approach used in the invention described in the OP, with a pre-chilled slug used to set up a temporary gradient. In that case you could lose too much body heat to warming up the cold mass. But even without pre-chilling (i.e. in the steady-state case), while you may be right that a TE device necessarily slows down the conduction of heat across a gradient, I don't think it follows that any such 'suit' would result in overheating rather than hypothermia. Consider 2 people in a 70-degree (F) room. One wearing light clothing, and the other sitting in a tub of water. The water in the tub will equilibrate at somewhere warmer than 70 degrees, and the person will become hypothermic, since the capacity of their 'suit' to conduct heat to the environment is greater than the ability of their body to generate heat to compensate. You could have a TE device that extracted power from the gradient between the warm water and the room without saving the person from hypothermia. That is to say, I think we're both wrong :): a well-designed suit could keep the amount of heat transferred within the range of the needs and capabilities of the human body. (Fun trivia: according to Murray Hamlet, a US Army cold weather injury researcher, the leading cause of 'morgue wakeups' in the US is drug users who pass out in the bath, become severely hypothermic, and are wrongly declared dead).