business under autonomous AI control if the potential profit is large enough.U.S. law already allows for the possibility ofAIsbeingconferredalegal personality, by putting them in charge of alimited liability company. A humanmayregister an LLC,enter into an operating agreement specifying that the LLC will takeactions as determined by the AI, and then withdraw from the LLC (Bayern 2015).The result is anautonomously actinglegal personalitywithno human supervisionor control. AI–controlled companies can also be created in variousnon–U.S. jurisdictions; restrictions such as ones forbidding corporations from having no owners can largely be circumvented by trickssuch as having networks of corporations that own each other (LoPucki 2017). A possible startup strategy would be for someone todevelop a number ofAI systems, give them some initial endowment of resources, and then set them off incontrol of theirown corporations. Thiswould risk onlytheinitial resources,while promisingwhatever profitsthecorporation might earn if successful. To the extent thatAI–controlled companieswere successful inundermining more established companies, theywouldpressurethosecompaniesto transfer control toautonomous AI systems as well.
5.2.2. Voluntarily released for purposes of criminal profit or terrorismLoPucki (2017) argues thatif a human creates an autonomous agentwith ageneral goalsuch as“optimizingprofit”,andthat agentthenindependentlydecides to e.g.commit a crimefor the sake ofachieving the goal, prosecutors may then be unable to convict the human for the crime and can at most prosecute for the lessercharge of reckless initiation. LoPucki holds thatthis “accountability gap”,among other reasons,assuresthat humanswill create AI–run corporations.Furthermore, LoPucki (2017) holds thatsuch “algorithmic entities” could be createdanonymouslyand thatthem having a legalpersonality would give thema number of legalrights,such asbeing able to“buyand leaserealproperty,contractwith legitimatebusinesses,open abankaccount,suetoenforceitsrights,orbuy stuffon Amazonand have itshipped”.If an algorithmic entity was createdfor a purpose such as funding or carrying out acts ofterrorism, it would befree fromsocialpressureorthreatsto human controllers:In deciding to attempt a coup, bomb a restaurant, or assemble an armed group to attack a shopping center,a human–controlled entity puts the lives of its human controllers at risk. The same decisions onbehalf of anAErisknothingbuttheresourcestheAEspends inplanningand execution. (LoPucki2017)While most terrorist groups would stop short of intentionally destroying the world, thus posing at most a catastrophic risk, not all ofthem necessarily would. In particular, ecoterrorists who believe that humanity is a net harm to the planet, and religious terroristswho believe that the world needs to be destroyed in order to be saved,could haveaninterest incausinghumanextinction(Torres 2016, 2017, chap 4.).
Upload your study docs or become a
Course Hero member to access this document
Upload your study docs or become a
Course Hero member to access this document
End of preview. Want to read all 75 pages?
Upload your study docs or become a
Course Hero member to access this document
