The Hawking Protocol.
- Notes on surviving the coming intelligent machines.
email: eldrasATlondon.com
The Hawking Protocol is a LondonAIClub specification protocol for the containment of runaway artificial intelligence.
"In contrast with our intellect, computers double their performance every 18 months, so the danger is real that they could develop intelligence and take over the world." Prof Stephen Hawking August 27th 2001 Focus.
The danger is traditional inaction resulting in extinction.
The Collingridge dilemma:
" impacts cannot be easily predicted until the technology is extensively developed and widely used, by which time it is hard to make it safe."
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
"The survival of man depends on the early construction of an ultra-intelligent machine." I J Good 1963-4.
**********************************************************
Background and Contacts.
An increasing number of people are warning about intelligent machines. The late Oswald Minter, a collateral descendant of Sir Isaac Newton, had warned us for decades about the advent of machines though London Artificial Intelligence Club members regarded him with amusement.
Alan Turing had noted in 1951 'Intelligent Machinery':
"once the machine thinking method has started, it would not take long to outstrip our feeble powers. ... At some stage therefore we should have to expect the machines to take control,"
Famous high techies and philosophers early-foresighted existential risks.
1993: Vernor Vinge in his NASA address said within 30 years the human era will be over: The Coming Technological Singularity: How to Survive in the Post-Human Era,
Why The Future Doesn't Need Us "Our most powerful 21st-century technologies - robotics, genetic engineering, and nanotech - are threatening to make humans an endangered species."
Bill Joy founder of Sun Microsystems.
A.I. expert Hugo de Garis has noted that AI may eliminate the human race: humans would be powerless to stop them because of a technological singularity.
Solutions to the threat of artificial machine intelligence:
1. Merge with it: (Update human beings faster than machines).
2. Build and contain Superintelligence: (to neutralize all risks).
The AGI conferences are for people trying actively to build Superintelligence, and some attention is given to dangers:
Conferences on Artificial General Intelligence.
Solutions like shields/barriers against it cannot logically work, as they would assume greater than or equal to Superintelligence defence systems.
The Hawking Protocol, is a system of protocols for building safe A.I. from philosophy through technical specifications of various machine and artificial systems to eliminate its dangers absolutely.
A theory for this was not achieved until 2007 under the B.I.D.I. at London A.I. Club, despite building our first prototype accelerating intelligence in 2001. Set up in 1999 to test the proposition A.I. was impossible to build:
The Club concluded (Superintelligence) was not impossible to build.
The Club concluded a restraint of intelligence amplification impossible.
The Club attended The Royal Society and warned of the dangers of A.I.
The Club attended AI@50 and warned that accelerating A.I. was not containable because of the logical proposition "A greater intelligence cannot be contained by a lesser intelligence."
The Club lobbied and warned governments of the UK, EU and the UN of the dangers of A.I.
The Future of Humanities Institute at Oxford and others like the The Cambridge Project for Existential Risk are seeking ways to avoid human extinction from Artificial Intelligence.
"Only a full solution to the problem of AI Friendliness would be guaranteed to work....However, I think that a full solution to the problem of AI Friendliness is almost surely impossible..." Ben Goertzel (to me) 2012.
The Association for the Advancement of Artificial Intelligence deals with risk as A.I. Ethics, and considers mitigation of risk in emerging machine and artificial intelligence systems.
Oxford Professor Nick Bostrom has written a paper analysing a date for the advent of Superintelligence
http://www.nickbostr...telligence.html
and has a book due "Superintelligence" in 2013 OUP.
Wemust soon race to prevent extinction by a breaking science that dwarfs runaway pathogens, nanotechnology and global warming.
The UK government commissioned a 100 scientist team to predict inteligent systems for 5, 10 and 20 years:
http://www.bis.gov.u...gnitive-systems
Note:
Cognitive Systems 2020 - European Foresight Platform
Centre for the Study of Existential Risk - University of Cambridge
Foresight Institute
Machine Intelligence Research Institute
Future of Humanity Institute - http://www.fhi.ox.ac.uk/ University of Oxford
@ which see:
Stuart Armstrong:
and the related paper:
http://www.aleph.se/...rs/oracleAI.pdf
" some ideas in our oracle paper http://www.nickbostr...pers/oracle.pdf
and some ideas from Roman Yampolskiy (see http://www.ingentaco...020001/art00014 https://singularity....Engineering.pdf ).
"Paul Christano had some very good ideas, that seem to be unpublished (to summarise one strand: if we have whole brain emulations, we can make AI safe in a specific way)." S.A.
The Machine Intelligence Research Institute
has been researching it since the 1990's, doesn't see containment as an option and is seeking to make sure it it friendly.
See also:
Lifeboat Foundation
Association for the Advancement of Artificial Intelligence (see it's ethics writings/publications and conference debates, eg http://www.aaai.org/...rticle/view/540)
Less Wrong has an article here which is important and has links to papers on risks and counter measures.:
http://lesswrong.com...l_risk_from_ai/
see also:
Global Catastrophic Risk Institute
It is hard to see how adequate safety can be built without a design for Superintelligence which only a few of us claim to have.
We should rush to a containable build of Superintelligence because of the accelerating dangers of technology and other existential risks eg asteroid strikes.
“If you will not fight for right when you can easily win without blood shed;
if you will not fight when your victory is sure and not too costly;
you may come to the moment when you will have to fight with all the odds against you and only a precarious chance of survival.
There may even be a worse case. You may have to fight when there is no hope of victory, because it is better to perish than to live as slaves.”
― Winston Churchill.
Link to old Kurzweil site on THE HAWKING PROTOCOL
http://www.kurzweila...D=18526#id18526
Edited by Innocent, 29 August 2013 - 10:23 AM.
