Another 0.6 BTC means AM has to generate more than 240K BTC in the future.
This looks like a fallacy to me.
It would be true if AM's only gains would be from mining, but they will come from selling to other users instead, so there's no hard limit on how can they earn.
Of course the users would be mining themselves with their chips, and your post would imply they won't break even... well, yes, might be true. Still, they might buy anyway, like they did so far so often.
Yes, most buyers are too optimistic in buying hardware but we cannot expect them never do calculation before buying. More and more buyers have learnt the lessons and becomes more realistic in calculating mining income. Therefore, AM can earn more than what their chips can mine, but not too much. If they are lucky, they may even double the BTC by selling due to the over-optimism of buyers , but it is still proportional to the number of BTC left (as long as tx fees can be ignored). Moreover, I've ignored the cost in my computing and AM will not ignore them in dividend distributing.
This makes sense, but you may be leaving out two relevant (ASIC Mining gear) market segments:
1) New Bitcoin investors that prefer to mine their coins instead of buying them (there are different reasons for this that have been discussed somewhere else)
2) "Corporate Miners" that need to stay on top of the game
Also, you should be very cautious when declaring the whole mining industry "not profitable at current difficulty levels" - as long as Bitcoin prices keep moving they have, it will be extremely difficult to prove this assertion. Looking at ROI in BTC alone can be an extremely misleading simplification.