Atrojan horse is a program that advertises that it does one thing, but really does something that’s malicious. for example, a computer chess program that deletes the contents of the user’s hard drive when the user loses the game is a trojan horse. how would the bell-lapadula model prevent an attacker that only has unclassified access to a system from obtaining top secret information even if the attacker is able to trick his boss (who has top secret clearance) to run a trojan horse?