The conception of evil arose from
alternate reality of religion, the intention was ofcourse betterment of
individual hence society, simultaneous evolved awareness with degradation of
hedonist world makes conceptions of alternate reality -conceived in medieval
times, inadequate and meaningless to deal with complexities and urgent emergent
realities of contemporary world. Evil was created as neat binary of good (which
extended to juvenile conception of heaven and hell) for ease of convenience, a
simplistic premise of moral for people to latch on to. In the uncertainties of
medieval times, it worked to create workable ethical standards and synergy of
dealing. You can deal with evil, or atleast equip yourself with an
understanding, if he or she is a human. A human despite flaws suffers from
human condition. Despite the megalomaniacal sickness and grandeur infected from
alternate realities of religion and market they will have to face existential doubts
and weaknesses that are too human. Now imagine an evil beyond human conception,
an evil which is no longer human, an evil that is not contained in the binary
of good or evil but carries complexities of good and evil. An evil that is
beyond Kafkaesque -a suffering of modern times that mix existential and human
condition into neglect, a bizarre and illogical world of total alienation and
utter horror. An evil so logical that it questions our hold on reason. This has
never happened before in human history, it is unprecedented. This is what is unravelling.
In the recently concluded G7 meet
in Italy Pope Francis addressed the group on dangers of AI. This is
significant, now I do have strong reservation on religion as also its
pernicious presence in contemporary world but I do find Mr. Francis interesting
and indeed a valuable and influential presence. So, whenever he speaks about
anything other than religion, I give it due importance to probe further. The
risk of autonomous weaponized AI is real. AI weapons that are trained and authorized to select their own targets and attack without further human authorization. It an evil that cannot be adjudicated
with human reason nor contained by human condition. It augments ‘conventional, nuclear and cyber capabilities in
ways that makes security relationships among rivals more challenging to predict
and maintain and conflicts more difficult to limit’. AI is set to become
indispensable part of defense system as it provides invaluable strategic
advantage and insights that humans have no access to. It has capacity to
identify patterns beyond human conception that even adversary did not plan, and
then, recommend most optimal dynamic strategy to counteract in realtime. The
simultaneous transfer of information to top decision makers in crisis situation
is matter of concern since these become black box decisions -not accessible to
human logic or reason, indecipherable to humans -that may leave them unsettled,
but nevertheless certain of its strategic advantage. AI is superior to human
reasoning with access to deepest pattern of reality in a way that is beyond
human capacity, indeed range of human thought. There are two immediate issues here
when non-human logic is introduced into military strategy; firstly, the
strategic advantage created cannot be halted since enemy will be perceived to have
formidable advantage, and secondly, since these capabilities are undecipherable
accountability cannot be fixed hence severely impinging democratic discourse space
and diplomactic negotiation imperatives. AI already has an upperhand over
humanity. The race for strategic AI advantaged formidable defence system has
already begun, and most likely is being used by IDF. There is a need for
international moratorium on destabilizing AI capabilities that has exceptional
destructive capacities. It is an arduous task since AI’s separate logic system
by its very nature is incalculable hence generates autonomy and is adaptable to
extremely volatile reality. Human oversight is essential before it spirals into
most optimal strategic advantage bypassing humanity. AI cannot be autonomous. Transformative
AI capabilities needs verifiable restraints and authorization for irretrievable
actions where human lives are at stake. This new emergent reality needs
entirely new vocabulary to deal.
