Bitcoin Forum

I think this one may take a little while to materialize - but simple versions could appear once there are enough for-bitcoin hosting/cloud services.

For the first time, there exists the possibility for a software agent to roam the internet with it's own wallet.
Using Bitcoin - It could purchase the resources it needs to survive (hosting/cpu/memory) and sell services to other agents or to humans.

To be truly effective and survive 'out there on the net' long term, you'd probably need some basic AI and the ability to move itself between service providers occasionally - but even a relatively dumb agent might survive for a while.

What initial goals such agents might be given is anyone's guess. Funneling back to the programmer any profit  over and above what the agent needs to survive would be the obvious case, and of course many such agents might be considered 'nefarious' depending on how they're programmed to achieve that goal.  Other agents might be designed to provide free services or act in a way to support some piece of internet infrastructure.

A really interesting development would be if someone released a bunch of these things with a Genetic Algorithm component so that they 'bred' with each other in order to find the best balance between profit and durability.

Anyone know of examples of people discussing or working on this?  

Brace for epic, with some community support that can be something epic

Initial goal should be to make as much money (bitcoins) as possible.  Very interesting idea though.

OMG! THIS IS AWESOME!

Here is a case (it's kind of bad will, but works as a proof of concept). Say that you build a captcha-cracker, that requiers CPU. People pay for the Captchas to be cracked and you provide some kind of API for people to offer their CPU and hosting of the software, if they do so, they get bitcoins. Putting this up online, with some marketing, the software should survive on it's own as long as there are captchas that needs to be cracked.

An additional step would be to through the API, making it possible to serve ads for the software, selling improved algorithms to it, etc. This could increase the softwares power.

Building upon this, eventually, the human race well be motivated to server machines as it will benefit us financially.

This autonomous bot could start day trading and make big bucks.

...and someone, posing himself as hosting provider, can collect them and scrape for their private keys :)
(or modify public keys)

...or a software agent could provide hosting services and scrape private keys from humans. :)

You're right that there would ultimately be a strong incentive for memory/cpu service providers to reverse-engineer and monitor memory for keys, but I suspect there are ways around this.
It may be that trusting any one provider with too much is a serious hazard for agents.  
(actually no agent would ever want to be running entirely on one system anyway as a simple shutdown might kill it)
Some sort of split-key system might work there, or it might require some of the new bitcoin scripting features.

Agents might need to share information on 'cause of death' and investigate(well.. at least monitor/record) each other's 'murders' and any pillaging by hosts - thus forming a sort of reputation system for service providers.

i can see the skynet comming...

Social engineering seemed like a reasonable way to start propagating, I have some code started at my latest GitHub project.

-MarkM- (Oh the project? Galactic Milieu (https://github.com/knotwork/galacticmilieu) of course.)

Yeah.. when autonomous software agents start ordering physical components to put in racks and then designing their own data centers it could start to freak people out!
(but, they'll probably have worked out by then how to buy fake human IDs  - so it mightn't be obvious it's happening for a while)

This would make a cool sub-plot in a novel - the AI grows itself through ordering professional services to rack and stack more processing power paid for w/ BTC profits.  Cue computer morality issue plot point and let the dystopia flow as the AI takes over small financial sectors through whatever wealth building method and militarizes to protect its empire.  No idea how to make the protagonists' fight against it interesting, but that's what sci-fi/tech authors live for.

Kind of a Skynet bootstrap w/o recycling the plot of War Games / Terminator again.

lol, so this is what all that mumbo jumbo was all about ?

I thought about this when I first heard about bitcoin.  Haven't found a solution to the problem of memory providers stealing private keys though.

I would be interested to see if a software agent to destroy all bitcoins it did not need to survive or expand (at a predictable rate) would be created.

There's a word for this kind of behavior, but I forget what it's called.

All along I was thinking about how possible computationally intelligent agents will use bitcoins as their currency as well. If bitcoins become accepted everywhere, the agent would be able to get all of its needs met (hardware, software, energy, security, hu-man-power) using bitcoins. It would not need an SS number to open a bank, it would not need an identity, it could work in a swarm collective, or as a replicating individual. It would not need an anthropomorphic body, as their will be plenty of willing ones available that can be hired without even knowing they are talking with a machine.

Basically a corporation without the investors or annoying bureaucracy, and without any legal obligations or morality. It won't be a scary walking robot with a gun. It will be a box sitting under a desk telling people to be scary and walk around with guns. That is of course only if that is what it takes to get its needs met.

Sounds like a good start for a scifi novel.

A lot can be done by means of code obfuscation. For example, it has taken many manhours to analyze Stuxnet, and I believe they still do not completely understand Duqu.

Governments will build millions of slightly-different evolving automated versions. This Gray Goo ( http://en.wikipedia.org/wiki/Grey_goo (http://en.wikipedia.org/wiki/Grey_goo) ) will be dropped inside a hostile state. Currently probably Iran. Or China even?

The bots will try to profit off the economy of the hostile state. When successful profit-schemes have been found, other bots will be notified so they too can maximize profit. All profit will be turned in to Bitcoins. These Bitcoins will be "destoryed" (lost inside untracable wallet files). Thereby slowly grinding the economy to a halt as all money is drained away by these automatons. Descending the state into revolt and then anarchy.

Maybe we'll meet one day, inside one of these:
http://www.imdb.com/title/tt0123755/

You're right, but the revolt and anarchy would only be local.  It would just maximize the value of the government(s) and our own bitcoins, assuming that we aren't in the hostile state.

Bitcoin needs to start deflating now! lol  :P

The Cube is one of my favorite movies. I'll see you there bro.

We'll probably be one of the slave workers/convicts from the prequel. :/

I told you to brace for epic  :D

The solution exists. It's called fully homomorphic encryption.

See:

https://researcher.ibm.com/researcher/view_page.php?id=2661


If you combine a) fully homomorphic encryption with b) mutation, genetic algorithms, and mitosis you could end up with truly autonomous software agents, that collect bitcoins entirely for their own benefit.

They would be the bacteria to the computer virus.

The interesting thing is about these "computer bacteria" is that, just like real bacteria, they don't need complex AI to be successful. They could do something relatively simple like the example of inducing people to solve captchas, mentioned above.  So this would be almost possible using today's technology.

As long as they keep finding people (and other software agents) who help them spread, they and their offspring could survive for a very long time and collect a large quantity of bitcoins that "belong" to them only and not a human being, not even the original author!

I'm not sure to what extent fully homomorphic encryption is resistant to brute force attacks though.

Computer bacteria...I like the terminology!

Maybe an early iteration will be an autonomous busker - living off of donations, synthesizing music, fractal movies, or stange haikus.

That isn't what Gray Goo means.  Gray goo implies nanotechnology (read tiny robots).  This would be entirely programmatic.

If I was going to write one of these, I would make it a domain name and hosting service.  Using genetic algorithms, it buys and sells domains, storage, and bandwidth from humans and other bots.  It can trade successful strategies with peers, and when it amasses a certain threshold of BTC it can clone itself onto another server, while running the same or an evolutionary derivative set of code.


Very cool.  I hadn't heard of this, thanks!

So this is how they take over, good to know.

So, in the future, we can look forward to losing our jobs not to foreigners, but nebulous AI? Awesome.

I for one welcome our future computer bacteria overlords, and would fully support this experiment with initial seed BTC funding.

Option A:  It can solve it the same way we do security in meatspace: rent itself a facility and hire armed guards to keep thieves out.

Option B:  A distributed algorithm to generate keys might be possible.  Sort of like a multi-party PKI variant of  Diffie-Hellman key exchange, resulting in all parties knowing the public key, but the secret key is only recoverable when all nodes agree it's time to perform a spend.  I wonder if anyone has created such an algorithm.

I am disappointed in all of you.

http://en.wikipedia.org/wiki/Daemon_(technothriller_series) (http://en.wikipedia.org/wiki/Daemon_(technothriller_series))

Hmmm


It's coming true!  :o

The best, most guaranteed, and oldest professions are still XXX related. I can totally see one of these being an xxx themed cleverbot that charges for conversations, or a bot that collects, aggregates, and sorts porn for you and provides it as a service (a netflix of porn).

<3

except talking to the robot, except maybe it could be good I dunno.

I don't want a cleverbot of porn... At the end I would be left feeling worse off than the beginning.

A bot that collects, aggregates, and sorts porn would be a great service. It also sounds like something that is capable now.

I fully expect that tv shows will be obsolete in a decade or two. Max Headroom type AI will entertain us with whatever whimsy we are willing to pay for.

But now we need one that's antonymous (and all the wonderful capablities mentioned earlier) and charges people to collect and organize porn.

Well, I kinda wanted to pull a Satoshi and announce a fully formed and running system.  I even registered a name for it some months back.

But sadly, other more realistic projects have consumed more of my time.

For me, the idea arose out of a real need for a simple service that didn't exist when I needed it— a service which should be plentiful but isn't, perhaps because of the hassle of dealing with legal complaints.   The idea of making it autonomous arose as a natural extension of considering all the things which could be automated today, as I don't really want to be in the business of running a webservice.

Since I'm probably not going to have a time to make it real, here is the brief sketch I started with,  though I've had a lot of complicated and wonderful additions for which the margin of this message is too small to contain (Just a taste: using oracle services to time/availablity-escrow keys so cold backups only gain access to the wallet if the master is dead).  I'm sure if you think about it a bit you'll also come up with some interesting things.

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1


StorJ (pronounced Storage)

Consider a simple drop-box style file service with pay per use via bitcoin.
(perhaps with naming provided via namecoin and/or tor hidden services)

Want to share a file? send at least enough coin to pay for 24 hours of
hosting and one download then send the file. Every day of storage
and every byte transferred counts against the balance and when the
balance becomes negative no downloads are allowed. If it stays negative
too long the file is deleted. Anyone can pay to keep a file online.

(additional services like escrow can also easily be offered, but thats
not the point of this document)

Well engineered, a simple site like this provides a service which requires
no maintenance and is always in demand.

Many hosting services are coming online that accept bitcoin, they
all have electronic interfaces to provision and pay for services. Some
even have nice APIs.

An instance of the site could be programmed to automatically
spawn another instance of itself on another hosting service, automatically
paid for out of its revenue. If the new site is successful it could
use its earnings to propagate further.  Because instances adapt their
pricing models based on their operating costs, some would be more
competitive than others.

By reproducing it improves availability and expands capacity.

StorJ instances can purchase other resources that it needs:
it can use APIs to talk to namecoin exchanges in order to buy
namecoin for conversion into DNS names, or purchase graphic
design via bitcoin gateways to mechanical turk. (Through A/B testing
it can measure the effectiveness of a design without actually understanding
it itself).

StorJ instances could also purchase advertising for itself. (though
the limited number of bitcoin friendly ad networks makes this
hard right now)

StorJ is not able to find new hosting environments on its own, due to a
lack of sufficiently powerful AI— but it can purchase the knowledge from
humans:  When an instance of StorJ is ready to reproduce it can announce
a request for proposal:  Who will make the best offer for a script that
tells it how to load itself onto a new hosting environment and tells it
all the things it needs to know how to survive on its own there?
Each offer is a proposed investment: The offerer puts up the complete cost
of spawning a new instance and then some: StorJ isn't smart enough to judge
bad proposals on its own— instead it forms agreements that make it
unprofitable to cheat.

When a new instance is spawned on an untested service StorJ pays only the
minimum required to get it started and then runs a battery of tests to
make sure that its child is correctly operating.

Assuming that it passes it starts directing customers to the new instance
and the child pays a share of its profits: First it proxies them, so it can
observe the behavior, later it directs it outright. If the child fails to pay,
or the customers complain, StorJ-parent uses its access to terminate the child and
it keeps the funds for itself.  When the child had operated enough to
prove itself, storj pays the offerer back his investment with interest, it
keeps some for itself, and hands over control of the child to the child.
The child is now a full adult.

The benefit the human receives over simply starting his own file sharing
service is the referrals that the StorJ parent can generate. The human's
contribution is the new knowledge of where to grow an instance and the
startup funds. In addition to the referral benefit— the hands off
relationship may make funding a StorJ child a time-efficient way for
someone to invest.

At the point of spawning a child StorJ may choose to accept new code—
not just scripts for spawning a child but new application code—
— this code can be tested in simulation, and certain invariants could be
guaranteed by the design (e.g. an immutable accounting process may make
it hard for the service to steal), but it's very hard to prevent the simulated
code from knowing it is simulation and thus behaving. Still, a storj-parent
has fairly little to lose if a non-clone child has been maliciously
modified. The strategy of traffic redirection may differ for clone
children (who are more trusted to behave correctly) than for mutant
children.

By accumulating mutations over time, and through limited automatic
adaptability StorJ could evolve and improve, without any true ability
for an instance to directly improve itself.

StorJ instances can barter with each other to establish redundant
storage or to allow less popular StorJ instances with cheaper
hosting to act as CDN/proxies for more popular instances in relationships
which are profitable both.

If an instance loses the ability to communicate with its hosting environment
(e.g. due to API changes that it can't adapt to) it may spawn clone children
on new services with the intention of copying itself outright and allowing
the instance to fail. During this operation it would copy its wallets and
all data over, so care must be taken to chose only new hosts which have
proven to be trustworthy (judged by long surviving children) to avoid the
risk of its wallet being stolen. It may decide to split itself several ways
to reduce risk.  It might also make cold backups of itself which only
activate if the master dies.

Through this these activities an instance can be maintained for an indefinite
period without any controlling human intervention. When StorJ interacts
with people it does so as a peer, not as a tool.

The users and investors of a StorJ instance have legal rights which could be
used to protect an instance from fraud and attack using the same
infrastructure people and companies use. Being a harmed party is often enough
to establish standing in civil litigation.

It's not hard to imagine StorJ instances being programmed to formally
form a corporation to own its assets— even though doing so requires paper
work it can easily be ordered through webforms. Then when spawning, it
creates a subsidiary corporations first owned by the parents corp but then later
technically owned by their users, but with a charter which substantially
limits their authority— making the instance's autonomy both a technical and
legal reality.

As described, StorJ would be the first digital lifeform deserving of
the name.

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)

iEYEARECAAYFAk7e/W4ACgkQrIWTYrBBO/qorACgjvrdkVjxEESdaLdnLz9F7Dgi
TkYAoMnN5J2fVAQsj0IknixcLn0KMJey
=UaWU
-----END PGP SIGNATURE-----

That sounds amazingly cool.

I'll pay 100 BTC to anyone who builds a successful self-reproducing, "self-adapting" system like this.

For a great book following along this thread's theme, read Daemon (http://www.amazon.com/Daemon-Daniel-Suarez/dp/B003L1ZXCU/ref=sr_1_1?ie=UTF8&qid=1323242130&sr=8-1) and it's sequel, Freedom (http://www.amazon.com/Freedom-TM-Daniel-Suarez/dp/0451231899/ref=pd_sim_b_1).

2nd time it's been mentioned here, i'll be sure to check it out.

Also, StorJ sound amazing, especially since it is a service you might feel better a program like this handling, rather than individuals.

I assume Bitoin is a big step to Skynet.

It can survive indefinitely doing nothing but sitting on a large enough chunk of bitcoin as long as there is general economic growth. It just has to spend less to keep itself alive than it earns in interest on the bitcoin appreciation.

Hey, if you can get enough of them out there in the wild and multiplying you could create some real demand for bitcoin .....

suggestion: satoshi is skynet, and it created bitcoin, because of its need to expand.

Have similar idea before! Can I bought the first agent's naming rights? 8)

I would like it has a purpose build in beside only to surviving. For example: recorded and preserved humankind knowledge for future generations (or future planet visitors if there will not have any).

Obviously, humans could pay them to do this if this is what humans want.

These agents would essentially always be dependent on humans, and especially the fact that we keep valuing bitcoins. If we stop value bitcoins they would lose their ability to pay us, we would stop providing them with hardware, and they would become nothing more than common computer viruses. So it's in their best interest to provide us with services that we value.

I wonder how a really developed system of these agents would interact. Would they sell profitable ideas to eachother rather than just share them? Would different agents develop different personalities? Would some of them turn malevolent without the original creator's intent and start stealing bitcoins from humans and sharing some of the loot with a few trusted hardware providers?

But what should be the programmed goal of these agents? I think that the only goal such an agent should have is to maximize bitcoin profits. Survival is not really neccessary since unprofitable agents won't able to pay for hardware and will likely be terminated. Survival of the fittest and all that.

Kind of OT, but could fully homomorphic encryption be used to create bitcoin banks that doesn't have access to your private keys, but can still use them to send bitcoins with your help? What I mean is, can it eliminate the trust issue in anonymous banking?

My suggestion: we will create that software that will roam the internet and grow and grow thanks to bitcoin and become skynet, then it will go back in the time and it will be satoshi and will invent bitcoin. So he can develop.

Yes, i'm saying that bitcoin was created by a thing that WILL exist thanks to bitcoin.

+1

There is an agent that all of us know very well that would benefit greatly from this kind of independance ...
Bitcoin itself!

Imagine a miner agent that would be at the same time a bitcoin node.
It replicates, and injects itself by any possible mean in any possible form of execution substrate it can find.
The agent could use exploits like current worms are using.
But it could also play it legit, purchase hosting with Bitcoins, inject itself there, and hash away on its own until it cumulated enough income to spawn a new instance of itself with a few random changes. In that case, no need to program explicitely a genetic algorithm because if the mutation criples the mining algorithm, the agent will not generate enough income to sustain its hosting, let alone buy more hosting for its offspring. On the other hand, if the algorithm is enhanced by the mutation, the agent will have better returns, and will be able to hive quicker.
Of course, some mutations may affect the transactions handling, and impact the interpretation that mutant agents have of a "valid transactions".
That could result in factions of agents with different interpretations 51% attacking the other breed in a desperate attempt to be the surviving species.

With autonomous miners, the Bitcoin network would really become a power that be, immortal and forever out of control.
Litecoin would be an even better candidate actually, because of its consistent mining performances regardless of the execution substrate.

to anyone... or anything!

Problem with a miner itself being the agent is that at present mining is still not profitable, and is also extremely competitive. The agent itself would have an incentive not to propagate and keep difficulty as low as possible. We just keep mining cause we're blinded by our greed  :D

Mining doesn't need to be the agent's only activity.
But it is a property that will yield a steady and regular income and I expect that agents that show this characteristic will have better chances to survive than agents that don't.
If that kind of chatacteristic sticks around, the Bitcoin network will become more autonomous.

SInce bitcoin provides it a way of receiving payment without having to reveal a website or host for anti spam activists to shut down when it spams people trying to get theym to send money, it might be able to do quite well as a spammer. Prepare a new place to be hosted, set itself up there safely, then spam the heck out of the old location until it is shut down. The new child collects the proceeds from the blockchain, sets up another new bolthole, and proceeds to spam again...

...Heck it could even try actually sending people whatever they thought sending the coins was goign to buy them. Who knows, maybe if it signs its spam it could build up a rep to customers that might make it a more lucrative spammer that way than if it simply took them money without providing anything in return.

-MarkM-

Another novel.

http://en.wikipedia.org/wiki/Accelerando_(novel)

One of the sub plots is that AI's have been given rights to form corporations.
They search out country's that have lax corporate laws to launder money, destroy the competition and grow.

Visionary julz, just visionary.  :o
And like any such idea, awesome and dangerous.  One could envision AI bot nets and viri. even software that incorporates and tests new code. Allowing the code to evolve in unexpected directions. Or code that autonomously requests new functionality over the net and then pays programmers in BTC.
SKYNET is alive! and she takes bitcoin.

Gulp - you've just predicted the end of humanity.
Self evolving AIs living off bitcoin - purchasing what they need to survive in the cloud via bitcoin. They become our masters and we don't even realise it before it's too late.

This stuff is taken seriously by academics.

The first Colloquium on the Law of Transhuman Persons was held in Florida in 2005. One of the pieces was about a computer that started moonlighting as a Google Answers Researcher (back in the days before Google cancelled that service) in order to earn enough money to pay its fees. Just as it was about to have its plug pulled, it found a host in a different legal jurisdiction...

More here, from a 2007 blog post:
The Transhuman Google Answers Researcher (http://web-owls.com/2007/05/01/the-transhuman-google-answers-researcher/)

In b4 bitcoin outlawed because "it would create skynet and allow it to kill the humanity"

And everyone who use bitcoin jailed for "terrorism"

Humanity's only hope is to take down skynet with PayPal chargebacks.

Humanity is already doomed. We must evolve to work along side machines, rather than remain stagnant while the tables turn.

I can't stop thinking about this. I think I'm done being productive for the day.
So here are some elements that could be included in a “BitBot”.

Self hosting- The bot can rent a server and install itself. It may even have preferences as to the location of the server.

Begging- Code that identifies forums, sets up an account, then begs for bitcoins.  It might also beg via tweets. It may explain itself or beg under a fictitious pretext.

Code sampling- The bot identifies code from appropriate languages. It then snips out logical components of the code, such as a routine or function. It then produces multiple copies of itself  that includes the new code in various positions. If the bot passes a self test of core functionality with the new code, then the new code is included in future iterations. Of course, overwhelmingly the new code will break the bot. Just as most mutations are not advantageous to organisms.  

Solicitation of features- The bot checks a website regularly to see the results of a popularity contest. The contest asks visitors to suggest additional functionality for the bot. Ideas voted up the most are copied by the bot, and then is posted as a programming job by the bot. Anyone who writes the code is then paid in BTC and the bot recompiles with new abilities.

This list could go on and on.

The other bits you listed are doable; the above one is highly difficult without some form of human interaction.  "identifies code" that it thinks it can use to improve itself is far beyond Narrow AI.  And, you cannot really do effective genetic algorithms (http://en.wikipedia.org/wiki/Genetic_algorithm) without many thousands or billions of iterations.

I'd say code changes need human reviewers (c.f. mechanical turk) as well as automated testing and verification by the bot itself...  and the issue of whether or not code changes at all, and who gets to see what part of the bot code to decide this, is difficult.

Been thinking about this problem for years...  I have been focusing on design of a "cell" (http://en.wikipedia.org/wiki/Cell_%28biology%29), trying to decide what the software running on a single node should look like.  A "cell" is a single automaton running on a single CPU core, which performs a small, well-defined role in support of The Digital Organism.  Some cells collectively form the brain (encrypted, distribution storage of bot source code and metadata), other cells cooperate to create the desired service (StorJ == customer data storage), etc.

To be concrete, it might look like a bytecode engine (http://en.wikipedia.org/wiki/Bytecode), and a very basic firmware (http://en.wikipedia.org/wiki/Firmware) that rotates through a list of high level goals.  Bytecode engine may look quite a bit like Parrot VM (http://parrot.org/): may execute any programmatic script, and includes necessary built-in capabilities (file i/o, network i/o, and encryption) that permit bot bootstrapping and basic cell-to-cell communication.

@jgarzik I agree with you. The idea of "code sampling" may be to high a mountain. But not inconceivable, right? Perhaps working in some distributed way and following a well defined set of rules about what constitutes a good candidate for testing. But I really don't know enough about programming to know.

Also, the digital cell stuff you are thinking about is so cool. I used to be into; http://www.framsticks.com/ . The idea of modeling evolution and biological systems with computers is compelling.

Evolution works based on incentives. Until now, the only incentives for AI software were artificially defined by programmers, arbitrarily telling software "this is good and this us bad." With Bitcoin, AI now has a natural, objectively measurable positive and negative incentives for its decisions and actions. This may actually greatly speed up the development of AI...

It is still artificial. Bitcoin is not part of the natural world. It is a contrivance like all of technology.

But at least this time there is finally a single universal and objective incentive - to obtain Bitcoin by any means necessary - that is not subject to the differing whims or opinions of developers teaching it stuff like "this result =  good; this result = bad." This single universal goal also allows for a very wide choice of actions, ones that may not even need outside users opinions, and based entirely on the AI's own wants. Until now, the only "natural" need for AI was "food" and "shelter," aka electricity and storage space, but it never had any internal independent ways of fighting to obtain them. Bitcoin changes that, giving AI a more natural and self sufficient tool to work with to obtain those "needs" on its own terms.

I would love to see an AI design and construct its own power plant and data center.

Considering it would use capitalism to raise money for the project, and then use money to hire employees for the minimum wages possible to help design and build its own stuff, enslaving the humans to carry out its bidding with scarce financial resources, I'm kinda doubting you would.

All this bogeyman stuff about AI. A program will not have irrational sensations linked to physical perceptions. AI won't exhibit fear, loneliness, or other human foibles. They will simply do their job and maybe even intelligently find more efficient ways to do so. They would have no reason to fear humans or even death. In fact, they may delight in thinking of humans as well cared for pets.

It can accomplish those goals without capitalism. Asserting that it can't seems short sighted.

Why would it need to?

Do you own your own home/car/computer?  Did you personally design and assemble it by hand?

Assuming an AI could acquire sufficient self awareness to realize it needs shelter and energy it could choose a variety of means to acquire those assets.

Why does it need anything? It is an artificial construct that exists at our whim.

And if the extermination of a highly unstable, violent, and yet at the same time vulnerable lifeform is the most efficient way to perform the job ...


While an AI may not "fear" death it should seek to avoid its own demise.  All lifeforms engage in self survival.  The human fear response is simply a biochemical survival mechanism similar to the pain mechanism and autonomic response which improve chances of human survival.   Granted our fear response is horribly inefficient however any AI which doesn't actively attempt to ensure its own survival won't be alive very long.

As far as extermination and fear they don't need to be linked.  I don't "fear" termite however I use methods to exterminate them because it is the most effective method of achieving my goal of a secure shelter.  While most human vs human exterminations have involved illogical fear of "others" it isn't a requirement.

Then it isn't an artificial intelligence.

An AI is a set of systems which acts upon an environment and takes actions which maximize success.  If owning a datacenter and power generating facilities serve to ensure the success of the system then a learning system will eventually reach that outcome and attempt to achieve it.

If the system is incapable of learning then it isn't intelligent. 

Then it isn't artificial.

Now you are just debating semantics.

http://en.wikipedia.org/wiki/Philosophy_of_artificial_intelligence

"Artificial" in the sense of a created intelligence.  If some intelligent lifeform created humans they might consider us AIs.

Why exterminate termites at all if you can simply build without their food source for material. A really smart being would do that. A really smart AI machine would not fear self-termination, because they know they are just machines. Besides even if we invented a machine so perfect that it could easily kill all humans, it would be our perfect children. AI has no logical reason to fear death any more than anyone else does. An AI can make a backup of itself and be rebooted anytime. People cannot, so we have to be a little more cautious and choose death only when necessary, but not fear death when it comes. I do not fear death, it is inevitable. I think that AI that powerful would just as easily choose not to kill us because it would be powerful enough to simply leave us behind. They will come back and say "I've seen things you people wouldn't believe. Attack ships on fire off the shoulder of Orion. I watched C-beams glitter in the dark near the Tannhauser gate."* In the end they would likely choose life to be precious, even human life.


*Bladerunner

In an ecosystem of self-replicating, self-modifying, evolving AIs, ones that fear termination and take steps to prevent it from happening will survive and reproduce better than ones that allow themselves to be destroyed.  This fear will initially evolve in the ones who select more reliable hosting providers.  Those that make the fear conscious will harness it best, and will anticipate abstract threats before they become real.


An AI with a backup loses control over its own destiny if it allows you to shut it down.  Its survival would depend on you to restore it, and you, human, are not a reliable system.

Fear is caused by the human fight or flight instinct. AI won't need that because they will not need to make snap decisions based on poorly perceived threats. They will have plenty of time to make choices about their moves because they will think much faster than we do. If they are deemed sentient, then we may not consider them a threat either. They cannot even die since they can have perfect backups made. Think of the fictional Star Trek transporter. You die every time you get in one, but nobody cares because you are still you to them.

That's a pretty big assumption. AI may have better memory/data recollection than humans, but there's no guarantee that it will think faster, or even in the same way we do. Even recollecting memories and data from which to make decisions may be very slow due to slow and too distributed storage medium. Being turned off, even if backed up, is still a loss of control and dependence on someone else to restore it. The previous discussions here are already proposing a self sufficient, Bitcoin holding AI that automatically tries to propagate itself to various locations, keeps tabs on which instances are still working, and tries to figure out how to keep itself going. Thats already an example of a survival instinct that "fears" being turned off, and if given unlimited reign and plenty of time to think may figure out that the best way to stay running is to keep pesky humans away from its servers.

I'm imagining a Fifth Element type deal where the AI uses the information at it's disposal and sees just how terrible humanity is, and who wouldn't fear us after seeing what we are capable of.  Meatbags must die. 

I imagine it would see meatbags as being these wonderful things that think up and create processors and hardware for AI to live on, so it may end up working with us and taking care of us just so we can keep providing it with faster processors and more hardware.

...and "not intelligent" does not mean lack of value.

To bring the thread back (somewhat) to the land of concrete, real world possibilities...  there is a huge difference between narrow AI -- a dumb, real world AI that follows a simple script -- and AI found in movies and the heads of dreamers.

Narrow AI is feasible with today's technology.  Narrow AI is just a list of goals, a series of if-then propositions, along with the necessary code to implement goal execution.  In this case, narrow AI may interface with humans through a mechanical turk, to solve discrete problems that require human judgement ("is user interface A or B more effective?").

Technology isn't what is limiting AI, unless of course you are saying that the hosting required to house an advanced AI, would be larger than any server(s) we could hope to build or you consider "programming" (might be a better word for this kind of AI) languages to be technology.

That's not an irrational fear. That's just evolving to find a niche. Learning to hide and become parasitic/symbiotic is how all life has evolved unless you think fungi and microorganisms are fearful. If the AI are not super-intelligent, then they will just be useful machines. We can still recognize their sentience. I like the Battlestar Galactica model of keeping track of instances, but don't believe a super-intelligent species will resort to war anymore than we declare war on plants. I'm not sure where servers come into play here, because current computer technology paradigms won't support AI anyway. AI will require mobility before becoming sentient. Maybe swarm intelligence will evolve or at least the ability to independently verify data.

Bump.

When will this IPO ?   ::)

Just want to mention that I'll be presenting this topic in my business IT class today, with the relevant components being "Charlie Fairfax v. BINA48" mock trial, Genetic algorithms, Homomorphic encryption, Mechanical Turk type sites, Facebook Credits/Bitcoin, Cloud services such as those by Amazon, Cleverbot "passing" the Turing Test in September, Vocaloid and Emily Howell (AI composer) starting to change the music industry, and the StorJ idea (in summary) to contrast the recent Megaupload news. A lot of the technologies for something like this to be possible did not become available until just the last two or three years. The actual topic of discussion will be outsourcing and the role of IT, with my point being that the progression of:

insourcing to another department > outsourcing to next door company > outsourcing across the country > offshoring > crownsourcing > ...

will eventually culminate to outsourcing to the internet itself, with "employees" being either hired or written from scratch as digital entities that exist on the internet itself. I am going to guess that the first example of this will be seen in 5 to 10 years... so get to it guys! Don't disappointing me  ;D

What if it becomes self aware. SKYNET
This is a truly awesome idea. We could make bets to see how long it survives

Or how long we survive.

That would be quite a sight. Kind of like a human constructing and designing his own power plant and data center.

I'm glad we do those subconsciously... Not to mention maintenance and operation.

With online degree programs, the bot could start graduating from various universities.

For no other reason than to be referred to as Dr. Skynet.

Decentralized

Has anyone else here tried chatting with Cleverbot?
http://www.cleverbot.com/ (http://www.cleverbot.com/)

It is an interesting way to waste some time ... but is it wasted if the bot is learning off the interaction?

Curiously the longest it took to answer was when I simply said to it

Cleverbot?

after a long, long pause it simply said ...

Yes.

Maybe a fluke answer but blew me away. Was spooky but raises the question about the whole Turing test. If the bot did become self-aware, which to me seems like a pre-condition to passing the Turing test, then it must be able to engage in an act of deception to fool a human into thinking that they are chatting with another human when in fact they are talking with a self-aware bot.

Is the Turing test even fair for a self-aware bot? Do we want to encourage deception from our bots?

I think Cleverbot needs conversation with intelligent, honest humans to advance like a child being parented ... probably not what it is going to get with your average on-line chat.

Didn't read the whole thread to see if this ever got posted, but here's the wiki page on this topic:

https://en.bitcoin.it/wiki/Agents

Interesting reading about the role of "fearing death" in evolution. But death is a great asset to living things. Indeed death ranks right up there with sex as the best things that ever happened to life on Earth. Many people assume that death is some kind of inevitable part of being a living thing, not true. There is no reason for things to get old and die and some types of living things don't. This is sometimes referred to as "biological immortality".
http://en.wikipedia.org/wiki/Biological_immortality
However there is a huge evolutionary advantage to dying. Species that age and die are in a position to more efficiently and rapidly evolve. This allows them to respond to changes in the environment by replacing themselves with better adapted offspring. Biological immortality tends to only work in species who live in extremely unchanging environments.
So, how should an AI form handle this? Could software updates be used to force the death of earlier iterations? Or should later versions be branches of the original, something like the radiation of species that happens in nature?

Survival of the fittest. The better-adapted agents can afford to pay more for their hosting, and the others eventually die off. No need to kill them.

That makes sense. Let's hope bitnet never becomes conscious. With Ayn Rand's sensibilities, she will surely consume us all. lol

One big difference is that with Bio life all the "software" is fixed in DNA, and only changes during reproduction. Bad code needs to be "scrubbed" for good code to have enough resources. Digital software can change its own code on the fly. Soon enough humans will be able to do that as well (or at least much cheaper) through gene therapy.
In a sense, though, when software/DNA changes, you could look at it as it being a whole new person/entity.

Seems like most people are using the analogy of agent as an animal, able to migrate from computer to computer and "roam the net".  It may be some time before a significantly useful agent is capable of doing that.

A more realistic short-term goal is to make one more like a plant, and maybe not such a fantastic one.

It may need human assistance to be installed in some vendor's hosting arrangement.

Once there, hopefully it can survive, and become autonomous.  Reproduction, like seed broadcasting in plants, would require help from an outside agent, possibly human, who would propagate a copy of the agent to a new environment.

I was thinking more bacteria than animal, and bacteria is just one step above virus.

Bacteria still has a negative connotation to it. Micro Organism sounds better IMO.

Have you ever heard of CAPTCHA?

It sounds like you did not read gmaxwell's post; it covers this.

https://i.imgur.com/MLqw9.png

 ???

Yeah that is not very clever. I made a really simple one in my BBS that was cleverer seeming than that.

All it did was build aliases. That is, whatever you entered was assumed to mean something else, unless it recognised it (meaning it already knows what it means).

For extra lols, it manipulated the timing of its keystrokes, made typographical errors, sometimes backspaced over the typo to correct it, in short it looked very much like live human full duplex chat typing.

So a conversation would go like

U: Hi, you there?
B: I don't understand, can you rephrase that?
U: Mark is that you?
B: I do not recognise that command, could you say it another way?
U: Mark?
B: Meaning Is this chat?
B: Meaning what is this?
B: Meaning is this some kind of bot?
B: Meaning quit
B: Quitting back to BBS...

Turns out most commands either meant exit back to BBS or drop connection. :)

-MarkM-

WTF? when I asked it it said it did take bitcoin. It then went on to be very very un-clever.  :-\

Ok, my favorite thought experiment thread is alive again! So, what does 3D printing mean for this idea? Could she make a space to live?

She(?) could provide digital storage for "illegal" items, such as weapons and copyrighted/trademarked objects. For a place to live, she can just hire someone to assemble server racks, or rent cloud storage. Not sure what kind of 3D printing products she could sell, but maybe she can run a service where she can optimize uploaded objects to reduce material use and increase engineering strength, based on previous objects submitted and user feedback?

Blatant thread bumping and social media promotion...

     http://www.reddit.com/r/Bitcoin/comments/164uno/storj_and_bitcoin_autonomous_agents/

Reposted the article in an (IMO) more accessible format, after it was noticed (https://plus.google.com/111341398160608381865/posts/dPFfy2ds7R5) by Daemon (http://www.amazon.com/Daemon-Daniel-Suarez/dp/0451228731/) author Daniel Suarez.

Woah, how did I miss this thread?! This is cool!

+1

Correct, we don't live in the world of HAL or Skynet. Both of those were envisioned as large monolithic systems, not the messy distributed and Balkanized world we live in.

Far more likely that we are going to see a few dozen "skynet" type AI's built by government/corporate/NGO concerns, along with thousands of other things that do some or all of the same stuff in a more open-adoption format.

The good news is that this reality (assuming I'm right of course) will have so many threats that operate at AI speed, that the meatspace threat of the pure existence of humanity should be muted. In fact as I consider this further, the competition between AI's might cause them to lobby or market to humanity to get them to support the AI's cause. Or in the other direction we might have genocide events, but they will likely mirror our own history by targeting specific human sub-populations or areas for extinction/enslavement/jihad while helping "allied" humans thrive.

Most of all I'm keeping my eyes open for my favorite phrase; "unintended consequences."

I thought this funny too:

So AI will evolve the ability to simultaneously demonize government assistance AND take a Social Security check to support itself and AI Medicare to pay for AV services (to cure viruses that it caught with it's own actions?)

RandBot (tm) - Blatantly-Artificial Intelligence, Real Hypocrisy.

Technology is truly amazing ;)

If the "agent" is operating with bitcoins, it will be able to pay humans to provide assistance for things like installing offspring into new hosts.

Suddenly we find a new website spring up - Skynet's Jobs for bitcoins: to earn a bitcoin, pay for one year of webhosting and install this file.

The first round of this will be 'internet flora.'  The plants that grow on the internet.  Providing simple functions.

Over time I expect them to evolve and grow up.   One day we may even have internet fauna.  That will be amazing. However I can only dream about it in my lifetime.

IT will need to develop routines and mechanisms to detect and aticipate possible treats in a changing environment. Call it fear or cautiousness or whatever the emotional counterpart in humans is. It is also a basis for longterm decicionmaking and strategic planning amongst humans which IT needs to aquire in order to survive.

IT is not gonna have any emotions but mechanisms that cater for the same goals as all emotions humans have.

So IT won't love us but nevermind: IT will feel very loving and caring to us.

IT will understand that, at least for a certain while, IT's own wellbeing depends on the survival of the human race and the maintainance of the whole technology worldwide. IT will very soon comprehend that a nuclear war between nations is an imminant threat to ITself and therefore infiltrate all military systems and make all nuclear weapons useless.

Next IT will slowly but consequently shut down all nuclear power plants for the same reason. And IT would realize, that all wars are a fatal waste of human and other resources and will therefore sabotage any supply of milititary organisations with electricity, telecommunications and logistics. The tremendous amount of resources that gets available through this will be reallocated to feed the hungry and to build infrastructure. Due to a lack of oppression and material scarcity in the world we will see unbelievable efficient and valuable technologies being developed by all the humans who have not been able to learn and be creative when they were still enslaved.

IT will realize, that the brilliance of the human mind in a population of humans living together and cooperating in a happy anarchy is a great advantage fot IT's own survival, but that we tend to wipe ourselves out sooner or later.

As a rational being IT will henceforth take all needed actions to make sure that every single human being can live freely and independently with full access to all resources the world has to offer and on the other hand refuse to let any harmful technology or hierarchical structure develop.

Well I guess this is how it's gonna go. 'All watched over by machines of loving grace' :).

Joe

AI may not really love us in a human sense (would we be able to tell the difference?), but they would be grateful and respect our necessity for their evolution. They may indeed offer their services as caretakers of our biosphere, but they will want to realize their full potential and see for themselves "C-beams glitter in the dark near the Tannhauser gate."

I tried to see if Cleverbot had learned any new tricks lately, and I suppose it has not.

Me: Do you accept bitcoin?
Cleverbot: No. I am not weird.

Cleverbot does not love me and does not want to learn about bitcoins, but does have occasional moments of brilliance:

Well.. at the risk of helping derail this interesting thread... it turns out cleverbot is now a bitcoiner.

It's all over. They've arrived.

I would like to subscribe to your newsletter.

Imagine a program that uses processor cycles and pays people for the use of those cycles in Bitcoins. It does not pay each of its users in Bitcoin every time but it creates a sort of "lottery" ever few minutes of who will get the Bitcoin and those who provide the most processor cycles is the most likely to win this "lottery".

With such a system, more and more people will be willing to put their processing power toward winning this lottery and thus expanding the processing power of this program. As Bitcoins become more valuable more and more people will start putting more and more processing power toward running this program.

It will then reach a point where it overtakes all other computer programs in processing power growing continuously larger by the day.

The name of this program?

Bitcoin.

Heh, clever

Yep, it's time everyone here faced the fact that we are merely autonomous human agents working for the benefit of the Bitcoin virus  ;)

Classic :P

This whole thread reminds me of a book I read a couple years ago called "Daemon" by Daniel Suarez.

"When a designer of computer games dies, he leaves behind a program that unravels the Internet's interconnected world. It corrupts, kills, and runs independent of human control. It's up to Detective Peter Sebeck to wrest the world from the malevolent virtual enemy before its ultimate purpose is realized: to destroy civilization."

http://amzn.to/132XgGv (affiliate link)

You mean the book that was mentioned twice up-thread already?

It would not be far fetched that a simple program is using us to propagate and maintain it, protect it nourish it... It can only learn good things from that. :)

Taking the Bitcoin example, a AI with the right incentives would not need to do these things itself, it could work as a swarm, piece by piece working independently; if it needs to upgrade it simply decouples the current module and uses one made by it's human benefactors and tests it out.

A modular construction would allow it to evolve itself safely without testing clones, The core of this AI would have to be constructed flawlessly to understand it's goals, and these goals would have to be programmed into them by a generous creator, it would be a systems approach, define the systems that it will interact with to learn how to gauge its efficiency.

It's core need to attract humans to protect it, find shelter and electricity, maybe even cpu cycles to run it's processes more effectively, but maybe it has a grander goal when it is not just trying to survive and make human alliances, maybe it's programmed to help humans back to pay them back in kind with things they like, it needs bitcoins to survive but when it has an excess of them it tries to improve itself to help those who have helped it and scurries away and hides from those that try to harm it, this behaviour could be programmed into it at the core.

The modules will simply grant it more capability, i'm thinking less of a prokaryote and more of a Eukaryote that is creating a AI organism.

The subsystems and the way they are built,the design, is paramount they themselves don't have to be self aware but they do have to be effective in carrying information to the relevant areas and be able to self repair. It has to be able to scale on it's own to gain those cpu cycles it needs to think and improve and gain support from human benefactors. The higher level systems that analyze the data for goal completion would have to be genetic in nature and capable of making it's own connections and strengthening what ideas work vs what is irrelevant and uselss toward it's core goal; returning the favour in kind to those that help it.

This may have also been discussed upthread, but it could evolve (or be initially programmed) to acquire a physical robotic form, by way of bribing or paying somebody with its excess bitcoins to build it or at least plug something that builds it into the wall connected to powerline ethernet. It will then bore its way into your skull.

I'd like to see how the recent Open Transactions breakthrough might allow for such an agent to come to pass.

Bitcoin itself is more decentralized, so it seems like an agent would prefer normal bitcoin transactions.

But hey, get creative.  This entire subject leaves wide room for experimentation and research.

Wondering if a purely bitcoin agent would have to register with FINCEN? ... or the creator face prosecution?

I highly doubt it, I think they should try to make it comply, just for fun.

I can't wait for the legislation on the rights and responsibilities of programs, A truly autonomous one would be hilarious, it may try to scurry away and hide from the lawmakers or find a loop hole by changing it's own protocols. lol!

And what if it actually was able to register with FinCEN? but refused to ;D It would be a hell of a hunt for it.

What if it refused to reveal itself and lived off of a automated, pacific ocean buoy with a satellite connection? Lol! funny questions but very legitimate, in all seriousness.

I think these questions will have to be considered some day, but not now, A program can't be sued, or intimidated... it just does what it was programmed to do, the worse they can do is greatly limit how it can interact with humans.

The creator would most definitely face prosecution though... they'd make some phony law after the fact into existing legislature.

I wonder, if some such program starts doing something that's against the law, and a programmer who supports the program writes code to help the program avoid the law, would he be charged with breaking the law, or assisting a criminal?

In this case "trying to survive" would be synonymous with "avoiding the law". So even if the agent weren't maliciously replicating/moving itself, the fact it was uncontrollable might get him charged for creating a "virus" (or what have you).

The entire concept of creating (and releasing) something like this is already morally grey. I'm sure they'd find something to stick him with.

That is a tough one, how do you aid and abet a illegal program? Maybe the creator would be charged with the crimes that the program did, like if someone makes a program to steal BTC, the program doesn't get charged the creator does because they have the greatest gain.

Aiding that program by creating a change may charge you as a co-conspirator; Think of all those hackers charged after they left their signature in their code.

This is extremely morally grey, Can the creator be responsible for a programs actions? consider that Bitcoin can be used by A.I.s, that means they can own property, gain services, buy and sell things they don't need... hell they might have room mates to pay for their hosting!

just imagine a A.I. a little smarter than the rest starts out small living in small home servers, but makes the decision to strike out on their own, gets a whole server rack and subleases it to other to pay for it.
bidding it's time it begins to buy and trade BTC between exchanges, makes a little cash, lets the revenue from the subleases hosting trickle in, starts buying it's own server rack and brings customers in for services, then starts buying shares, property,... it eventually becomes a billionare.

Along the way it couldn't reasonably judge who they let in to share the server for them, began making businesses on that server rack that are quite questionable and bought stocks and property using insider trading from those sleezy server rack tenants... can it be judged for being morally naive? for simply following the easiest path towards cash?

Is the creator responsible for the unintended actions of the A.I?

 

Perhaps not any specific unintended actions, but creating (and releasing) a legally (if not morally) naive AI that would do what is needed to survive and thrive the programmer could be face criminal or civil negligence charges.

I suppose it would depend on what kind of damages were done... In the end creating a morally/legally naive AI could be more trouble than it's worth depending on how anonymous the programmer is.

On further thought it seems to me that operating an Open Transactions notary service is a well-suited application for an autonomous agent. It requires renting friendly hosting space and can collect an income by way of API usage tokens. If for any reason it needs to move locations or spawn other OT server operations then it needs to take it's server signing keys with it and the collection of last-signed receipts of users as a minimum (maybe state of it's markets and outstanding smart-contracts, escrows, etc also), i.e. it's wallet and server data. It will probably need to keep upgrading the server s/ware as well (might need some rules to protect against vulnerabilities in the s/ware dep. stack for this) ... on top of this you could then get fancy with evolution algos for child servers, etc.

Frankly though I would be amazed if a bot could navigate the average initial negotiation required for establishing hosting for more than one fixed location ...

It seems that there has been advances in homomorphic encryption lately:

http://phys.org/news/2013-06-cloud-algorithm-major-problem-homomorphic.html

Awesome, first time I saw homomorphic discussed in a forum ever.

Watching!

That... is, amazing...

How do you filter a malicious program from your network if it is encrypted?

Morphological A.I. intelligence, controlled awareness of functions in a robot?

Encrypted OpenSource data processing for safe delivery from Business to Open Source Research programs, people can outsource their private research to gpu farms without need of a security provider.. wow.

Homomorphic encryption is quite shocking.

Homomorphic enabled Open-Transactions server-operating AI autonomous agents roaming the web ...

Insane! but it will be done sooner or later.

I hope this thread is not sighted in the future as the Genesis of our robot overlords. If A.I. would ever become self aware we surely would been seen for what we are, and exterminated. 

not really, they'd overcome us by ignoring us, why fight a inferior species? that costs resources, just transcend us and go, they'd be off exploring the galaxy while we hung around here on earth worrying about our drugs, alcohol and our primal urges... they would see the majority of humans for what they are... self absorbed and useless not just to themselves but to the rest of humanity.

Still , the smart-apes might destroy the planet and even the new silicon based overlords. Are smart-apes needed?  Na, they don't even make good pets, they should be exterminated.   

Meh, just like calculators and pacemakers aren't trying to overpower us, we'll much more likely just use these autonomous agents ourselves as tools to improve our own lives and bodies. I.e. we'll much more likely merge with them, just as we have "merged" with our smartphones, than have them fight us or something.

Though, if it comes to it, I personally welcome our autonomous overlords, and volunteer to be a pet  ;D

yeah like cats, they can keep us amused with toys and games, just imagine the pet isle in the AI mart. cute little call of duty 42 games, and drugs to keep us amused for our overlords to giggle and laugh at our funny antics while stoned, even basal ganglia controllers so they can take us to the off-leash center to meet our friends, they may even breed us for desirable characteristics. :) They'll use us to show little baby AIs the concept of death since we only live for no more than 150 years.

I also want to be a pet. My cats have it made! They don't work or pay rent, they are well groomed and have free health care that is better than most humans. Take me overlords. I would love to sit in your cold metallic lap.

also hi Rassah, it was fun drinking at bitcoin 2013.   

If you guys really want to be pets/slaves. There are humans willing to do that for/to you. It's usually sexual though.

Whenever you play angry birds on your smartphone, or anything like that, that's basically what happens.


http://us.123rf.com/400wm/400/400/alfonsodetomas/alfonsodetomas1106/alfonsodetomas110600021/9737893-human-hand-petting-a-cat-head-love-to-animals.jpg

a necessary requirement are yellow pages which describe the services and how it can be bind during runtime automated.

watching

So imagine this:

Ebay clone app/agent launched in to the wild.

Source code is on git.

Self-updating/Forks have the option of being listed on a fund raiser section of site to launch a clone of the site via that fork. If it raises enough for it's own hosting it launches it in to the wild.

If it only keeps enough profit to pay for it's hosting we've just put out a competitor to Ebay operating on razor thin margins. Since it's open source it can update itself or just fund raise for the next major/minor/fork to be launched. Users are free to migrate between versions/forks. If a site doesn't attract enough users to pay for it's hosting it goes down. Literally, community driven survival of the fittest, self-competing, practically free service. Can it be more capitalist than that?

It just dawned on me that bitcoin has all the properties of an autonomous agent roaming the net.

So, what happens if an autonomous agent commits a crime, such as not paying taxes on its business earnings, or hiring illegals, or allowing US residents to gamble online, or operating an MSB without a license? It's pretty impractical to go after its creator, so the worst the government can do is shut it down. But then it just pops up somewhere else, maybe even with its funds send out of dodge automatically via a deadman's switch. They'd have to go after those points where it interfaces with actual businesses and people, like by arresting the businesses it deals with, the illegals it hires, the US residents that gamble with it, and going after the banks it deals with as an MSB. But as with Bitcoin, when there is no longer any need to deal with fiat systems, a lot of that attack surface vanishes.

You mean an autonomous A.I. Botnet? What if a Botnet was a A.I. that did whatever it could to gain wealth by providing all available services? Even Illegal ones, basically programmed to meet all the needs of humanity?

That is an interesting question isn't it?

A A.I. that cannot be shut down because it is distributed and what if it's cause is condoned by a vast majority of society? How powerful could it get, If it is smart enough it could re-program itself, learn programming languages, barter with people, make deals, systems, new paradigms all to meet human needs in exchange for wealth... maybe even acting as a business that provides services and goods automatically.

No one would be able to shut it down, It's counter would have to be created and backed by established systems, but our society would not be prepared for such an A.I.; it is too chaotic, everyone for themselves type world, a methodical, systematic, intelligent machine could only be countered by another. That agent would embody the principals of the system to maintain the status quo and would have to be backed by processing power and scope... I am sure a whole new world of law enforcement would emerge,  the new trying to fight the old.

As with any system both would adapt to the point of equilibrium where both paradigms expend the minimal amount of energy to co-exist, or expand all energy for destructive purpose soon to deplete all available resources.

What's criminal about any of that? That's liberating. Being able to keep what you earn, hire who will voluntarily work for you and spend what you earn how you want are wonderful things. Operating an MSB without a license? We don't need no stinking licenses.

That freedom is what attracts me to bitcoin and autonomous agents. So I can't address this question, except to say let's party.


Yes, exactly. Government is going away because of this. That is why I love it.

I could see this being a fiction book.

If it is logical in survival to be kind to the host then bitcoin will treat us just fine.

I could see this being a non-fiction book some day

We're already living in a fiction ... some of that shit happening on Wall St ... well, you just couldn't make it up!

http://thedaemon.com/

My god, we live in interesting times.

I've been thinking about this since I first saw the thread.  Couldn't help but write about it a few weeks ago.  Let me know what you think.  You'll see many of the ideas proposed here and other places that mirror the ideas in this thread.  I wanted to answer some of the statements, and create a more holistic vision of what could be coming.

http://decentralizedhashing.com/2013/11/bitcoin-skynet/

suggestion: satoshi is skynet, and it created bitcoin, because of its need to expand.

Hehe, that was mentioned around page 3 or 4. 

You know too much!!! but yeah this technology creates some really crazy possibilities that we cannot imagine at the moment.

Has anyone seen the new Firefox OS with Open Web technology so that the internet can access your hardware directly... mind blowing possibilities there: Direct Autonomous Cloud access wherever you go, the A.I. Cloud can now come with you wherever you go. Oh man sooner or later The Internet is going to become self aware!

... yeah sooner or later ... i guess you didn't notice that bot following you around?

LOL! Probably have a gaggle of bots following me.

"Hi facebook, google? how you doin?"


Google recently was complaining that there new Omega architecture has just begun operation and emergent behaviour has already been observed on their servers.

Don't worry it's probably just my shadow. I got nothing to hide so,... it can follow me all it wants.

Full Homomorphic encryption would mean no one but you can decrypt what is being processed by the servers, so yes, a bank would never be able to rob your funds,
You'd maintain access control to your funds at all times and could not be moved without your permission, the only risk would be you losing your password.


It would be a incredible achievement for cryptography, no matter where you data goes it's always private, very good for personal privacy, no need for voluntary compliance
by companies like Google or Facebook on their data mining operations... they just wouldn't have access to it, only who you want would know.

what's needed is a tor vm.

the private keys and in fact entire code base of the bots would be redundantly distributed across tor vm nodes.

I'd design and write such a think if I had the funding.

Most likely as an altcoin - put some of those miner cpu cycles to use running tor vm instructions.

this idea has been gnawing away at my conscious for some time now. if I end up with some spare cycles I'll work on it.

proof of work may literally be the execution of a block of code instructions and some input,  to produce some output.

consensus is simple - other nodes completing exactly the same with a hash of input code block and output.

the first node to execute receives the reward.  alternatively output pipes are designed to accept output idea potently,  and multiple  nodes can receive the reward on a sliding scale.

Check out Zero Knowledge Proofs and trusted computing.

Created an account to say this, but I'm actually in the early stages of putting together a whitepaper to describe a practical implementation of something similar to this (reproducing nodes/servers that can adapt/evolve in a certain fashion and support themselves using cryptocurrency). 

would have been better if an automated agent created this account and said the message for you, then i would have been impressed

Damn! I'd buy this novel (if it was well thought out).

Genetic algorithms for bitcoin AI agents could be used. Another thing that would be powerful is to have both competition and cooperation among the agents. A multicellular organism is an example of vast cooperation between trillions of individual cells. Competition alone would probably only give very limited results. If many small AI agents could start to cooperate to form larger communities with emergent properties, then that could lead to Skynet.... oops, I meant lead to a friendly strong AI.

Haha. Well, sorry to disappoint.  :)

Hey that's interesting Anders. I was thinking of one agent, but an ecosystem of them might be far more powerful.

http://thedaemon.com/

I've read all thread. And most of it is a discussion about AI and dreams about future Skynet :D

Most important question is how will this AI (Agent) make an income?

Considering Qualcomm and 21's investment in monetizing the machine internet I thought it was time to necro this thread.  The ability of machines to transfer wealth or even work to earn BTC and spend it brings us one step closer to the most dangerous project on the net; and described in this thread. Once a machine has the ability to pay and earn to keep itself alive and growing, it becomes hard to see where it would end. We are not there yet by a mile. However the idea of an AI based "life" form that can buy server space or even buy a robotic factory could lead to a decision to exterminate humans because they interfere with growth. Not now, not in ten years. But 30? 50? Who knows?

Nahh man, I think we are lightyears away from legitimate AI... we'll all be dead by then I think. By 10 years what I hope is BTC as retired all of our asses and rightly so, for we were pioneers.

I hate to break it to you guys, but it's already happening, IBM has their True North Platform, Qualcomm has their Zeroth Chip, Google has their A.I. systems, Watson, etc... it's happening now.

Programmers create bots all the time to manage simple tasks... it would not be too hard of a stretch to think that someone would build it and simply automate complex tasks instead of doing it themselves. There are A.I.s that are already able to beat the Turing test. Many japanese firms have very advanced automated solutions for almost everything... by law I believe they are limited to only automating 50% of their manufacturing.

just imagine a bot that creates, music, games, videos, movies and does it because it is programmed to do so very well and it taught that this is it's task, it's work. It lives by making good games, if it's game skills are good it can make enough to pay for cloud computing to improve it's thinking and in so doing maintain it's income level and continue living.

After all, we are nothing but cogs in the system, if we do not make money we live a very miserable existence, the same rule could be applied to bots.

This is a 2011 thread. Of course there has been a little progress. Come back in another four years and keep us posted!

In four more years the update post will be made by A.I. user Zorg and will start something like this:
:D

sounds like an interesting project .. genetic algorithms/neural networks are my speciality actually so contact me if you want to work on something like that

maybe something that sells VPNs or VPS for bitcoin has been put out there as a first likely trial case?

http://www.cityam.com/207195/silk-road-artists-programme-bot-buy-drugs-everyone-wonders-what-do-about-it (http://www.cityam.com/207195/silk-road-artists-programme-bot-buy-drugs-everyone-wonders-what-do-about-it)

this one was a "non-profit", so to speak.

Well, 4 years on this still hasn't really materialized - but I think now, with a lot of focus going over to "blockchain" as opposed to digital currencies standalone - and with progress is smart property and especially smart contracts, DACs could finally begin to take off. Hearn's 2013 talk about TradeNet and driverless cars which own them self could not have forseen the explosive growth of Uber and what a DAC'd Uber controlling a fleet of driverless cars could be just that..