11 The Self-Organizing Society
227
diverse interests and goals. These digital assistants could harness the potentials
of artificial and collective intelligence.
• Diversity is also crucial for distributed collective intelligence, which is better
suited to respond to the combinatorial complexity of our world than an artificial
superintelligent system.
10
11.1 Cybernetic Society Versus Synergetic Society: Why
Top-Down Control Will Fail
We have seen that, after the automation of production and the invention of self-driving
cars, the automation of society is next. However, there are two kinds of automation:
centralized top-down control and bottom-up self-organization based on distributed
control. The first option, corresponding to a technocratic solution, might be called a
cybernetic society, while the second one might be called a synergetic society, as it
builds on the local coordination of autonomous processes and on catalyzing mutual
benefits.
New kinds of information systems will certainly allow us to manage the world
more successfully, but we must be careful about the approach we take. Would a
cybernetic society work, or would a synergetic society be better? It turns out that,
for a number of reasons, it would be impossible for a supercomputer to optimize the
world in real-time. The attempt to create a data-driven “crystal ball” to predict the
future and a “magic wand” to impose the desired changes would sooner or later fail.
While digital tools, which collect a large fraction of the world’s data, can certainly
be powerful, they would not work reliably.
Therefore, a centrally controlled cybernetic society would sometimes make
mistakes. But as a powerful information tool might have a large-scale systemic influence, a single mistake could be highly destructive. For example, imagine powerful
digital tools to get in the hands of a misguided group of individuals or a criminal organization. This could easily create a despotic regime. Therefore, the more powerful an
information systems is, the greater are the safety measures needed to protect citizens
and companies from potential harm.
11 Today, however, no information system seems
to be a hundred percent secure. Most companies and public institutions have been
hacked already.
Surprisingly, not even a “benevolent dictator” or “wise king” with the very best
intentions and all the data and technology in the world could make optimal decisions.
First, although computational power grows exponentially, the data volume grows
even faster. Therefore, no single person, company or institution will ever be able
to optimize our rapidly changing world in real-time. Supercomputers cannot even
10 As we know, intellectual discourse can be a very effective way of producing new insights and
knowledge.
11 This calls for a suitable combination of encryption, decentralization, transparency, participation,
reputation systems, community moderation mechanisms, and legal protection.
227
diverse interests and goals. These digital assistants could harness the potentials
of artificial and collective intelligence.
• Diversity is also crucial for distributed collective intelligence, which is better
suited to respond to the combinatorial complexity of our world than an artificial
superintelligent system.
10
11.1 Cybernetic Society Versus Synergetic Society: Why
Top-Down Control Will Fail
We have seen that, after the automation of production and the invention of self-driving
cars, the automation of society is next. However, there are two kinds of automation:
centralized top-down control and bottom-up self-organization based on distributed
control. The first option, corresponding to a technocratic solution, might be called a
cybernetic society, while the second one might be called a synergetic society, as it
builds on the local coordination of autonomous processes and on catalyzing mutual
benefits.
New kinds of information systems will certainly allow us to manage the world
more successfully, but we must be careful about the approach we take. Would a
cybernetic society work, or would a synergetic society be better? It turns out that,
for a number of reasons, it would be impossible for a supercomputer to optimize the
world in real-time. The attempt to create a data-driven “crystal ball” to predict the
future and a “magic wand” to impose the desired changes would sooner or later fail.
While digital tools, which collect a large fraction of the world’s data, can certainly
be powerful, they would not work reliably.
Therefore, a centrally controlled cybernetic society would sometimes make
mistakes. But as a powerful information tool might have a large-scale systemic influence, a single mistake could be highly destructive. For example, imagine powerful
digital tools to get in the hands of a misguided group of individuals or a criminal organization. This could easily create a despotic regime. Therefore, the more powerful an
information systems is, the greater are the safety measures needed to protect citizens
and companies from potential harm.
11 Today, however, no information system seems
to be a hundred percent secure. Most companies and public institutions have been
hacked already.
Surprisingly, not even a “benevolent dictator” or “wise king” with the very best
intentions and all the data and technology in the world could make optimal decisions.
First, although computational power grows exponentially, the data volume grows
even faster. Therefore, no single person, company or institution will ever be able
to optimize our rapidly changing world in real-time. Supercomputers cannot even
10 As we know, intellectual discourse can be a very effective way of producing new insights and
knowledge.
11 This calls for a suitable combination of encryption, decentralization, transparency, participation,
reputation systems, community moderation mechanisms, and legal protection.