Information Technology Reference
In-Depth Information
membrane, blood group antigens on red cells surfaces, and other individual
biochemical signatures other than genetic differences (which are not
directly accessible to most nanorobot species). An alternate method for
uniquely identifying individual people is to insert an artificial gene into all
cells of each person from birth whose sole purpose is to cause cells to
synthesize a harmless simple chemical marker, perhaps a digitally encoded
short carbohydrate chain [73]—even a small hexasaccharide can encode
B
10
12
unique combinations [74]—that is released into the bloodstream
and is readily detectable and decoded by theater-restricted medical
nanorobots. The deployment of such self-renewable in vivo fiducial
chemical markers could serve as a convenient substitute for less-intrusive
phenotypic measurements but might be regarded as controversial much
like the 1999 USPTO-granted patent for barcoding humans [75].
15.4.4. Safety Protocols
Safety engineering involves making sure things do not fail in the presence of
random faults. As Dorner [76] notes that
learning theory tells us [that] breaking safety rules is usually reinforced,
which is to say, it pays off. Its immediate consequence is only that the
violator is rid of the encumbrance the rules impose and can act more freely.
Safety rules are usually devised in such a way that a violator will not be
instantly blown sky high, injured, or harmed in any other way but will
instead find that his life is made easier. The positive consequences of
violating safety rules reinforce our tendency to violate them, so the like-
lihood of a disaster increases. And when one does in fact occur, the violator
of safety rules may not have another chance to modify his behavior in the
future.
Since the risk of harm is great if medical nanorobots are misused, future designers
should try to make it as hard as possible to disable the safety features. To the
greatest degree practical, these features should be permanently embedded in
hardware to minimize the probability of circumvention.
Nanorobot control systems should employ failsafe designs which may
incorporate parallelism (dividing tasks among a large number of simple systems),
specialization (individual systems optimized for particular tasks), and redundancy
(comparing the output of multiple systems to improve reliability of the results).
These built-in safeguards are then enhanced by the use of safety protocols which
are active device behaviors designed to further enhance and reinforce safety.
Failsafe designs must ensure that even a total system failure will not lead to death
or serious injury of the patient. Safety protocols should be able to recognize
various catastrophic internal failure states including compromised physical
structure or software/data corruption, necessitating localized, intermediate, or
even whole system shutdown, with entry into a safe-harbor mode that may permit
Search WWH ::
Custom Search