Companies are a form of Artificial Intelligence

[This article is an excerpt from a comment I made on a Reddit thread, but I thought it was interesting enough to be it’s own post.  Original thread.]

Normally we think about AI as what the MIT guys started working on way-back-when, or else from like Neuromancer type stories where there are computer software that becomes cognizant and stuff.

Another term that’s been used in a similar but less different vein is Artificial Life, which removes the Intelligent factor and just basically says “a non-DNA evolutionary based entity” (like us, dogs or amoebas) that shows characteristics of living, usually like a good number of the 7 old life-signs: Movement, Respiration, Sensitivity, Growth, Reproduction, Excretion, Nutrition.

When I look at how organizations work (companies, governments, other organizations (Triple H, Red Cross)) they have many of the same signs that regular life has.

• Movement: Government will occupy new spaces with their military, and in the old days tribes would move with the herds.  Companies will relocate to a different state to get better taxation rates.
• Respiration: This is actually just a molecular gas exchange version of Nutrition, as we use the oxygen for energy and release CO2 as Excretion.  Artificial Life shouldnt be expected to have separate Nutrition and Respiration, since these are kinda arbitrary based on us and our kind.
• Sensitivity: Bad press because of a product?  Company releases a recall to minimize damages, reacting to their environment.
• Reproduction: Greenpeace will have splinter groups that have a new charter based on the old charter.
• Growth: Google doubled in company size, year after year until they hit about 50,000 total (employees/contractors).  All organisms hit a maturation size where for the purpose of that organism it becomes harder to grow.
• Excretion: Products and services could be one kind of “output” from an Artificial Life.  Carbon being released or chemicals being left over after manufacturing could be another.  Reams of paper being used up could be another.
• Nutrition: These types of Artificial Life seem to use money in the forms of revenues or taxes to sustain themselves, through hiring their employees/actors to perform the required tasks.  They also require raw resources such as components that become computers that are used, paper that is created, carpets, office space, land.  Since they arent DNA based, what it takes to feed them has a different tilt to it.

So that’s the Artificial Life component.  The biggest thing is that they have a life independent of the people that create them.  They can outlive their founders (Disney), they can replace their founders (Yahoo).  They can change many times and still be seen as essentially the same entity (Great Britain: Monarchy -> Magna Carta -> Present).

The Artificial Intelligence comes from how they interact with each other, and how they change over time.  You could have all the above characteristics and still not be intelligent.  But given the above A.L. definitions, and looking at say Intel or Lockheed Martin, they are producing some very intricately designed goods, and they improve these goods over time, and change their direction based on competition, the environment (what people want), and levels of funding (feast or famine, active or hibernate).   More importantly I think, none of the individual actors could produce these goods on their own, or even fully explain them all at the level they were designed, which means that the organization is capable of more intelligent actions than any of it’s actors.

Seen as an entity in their own they appear as independent operators, and in fact the rules that internally govern organizations change over time as well, so in the same “life span” they are evolving.  The Ford today is far and away different from the Ford when it was established, yet it is the same entity.

That was a long way of getting to this, but what I meant by “process based AI” is that it is AI that is created from processes, such as legal arrangements (incorporation), and hiring contracts (to retain employees/actors), and departmental goals (to retain the properly directed and skilled actors), intra-company communication (so marketing can tell engineering what to build) and extra-company communication (sales, support, tax collectors).

In this way I think they are AI, that are essentially immortal, and min/max for their own survival, using people as tools.  They arent fully sentient (thinking about themselves thinking about themselves), but they act intelligently, and they will adapt to their environments (laying off employees, offshoaring) to improve their ability to live, and avoid ending their lives.

Very few organizations are ever shut down, that dont just run out of resources.  They have a sort of “instinct to live” in the same way that we do.  They have gating factors such as if a CEO is taking the company towards it’s destruction, the Board will meet and fire the CEO and replace them with a new CEO that they direct to not take the previous course.  Yahoo is a recent example of this happening several times in a row as the entity tries to adapt to it’s new environment.