A lifetime learning to be a 9th Dan master at go perhaps? Building on the back of thousands of years of human knowledge and wisdom? Demolished in hours.… I still look at the game and it looks incredibly abstract!!
Don’t get my wrong I am really concerned, I just consider the danger much closer than others, but also more soluble if we look at the right problem and ask the right questions.
“Skeptical about the framework” I do not agree with. Indeed it seems a useful model for how we as humans are. We become expert to varying degrees at a range of tasks or services through training—as we get in a car we turn on our “driving services” module (and sub modules) for example. And then underlying and separately we have our unconscious which drives the majority of our motivations as a “free agent”—our mammalian brain—which drives our socialising and norming actions, and then underneath that our limbic brain which deals with emotions like fear and status which in my experience are the things that “move the money” if they are encouraged.
It does not seem to me we are particularly “generally intelligent”. Put in a completely unfamiliar setting without all the tools that now prop us up, we will struggle far more than a species already familiar in that environment.
The intelligent agent approach to me takes the debate in the wrong direction, and most concerningly dramatically understates the near and present danger of utility maximising services (“this is not superintelligence”), such as this example discussed by Yuval Noah Harari and Tristan Harris.
Yes—we have increasingly powerful utility maximisers already and they are in many applications increasingly dangerous.
6 describes the AGI as a “species”—services are not a species, agents are a species. 4 and 5 as written describe the AGI as an agent—surely once the AGI is described as an “it” that is doing something certainly sounds like an independent agent to me. A service and an agent are fundamentally different in nature, they are not just a different view, as the outcome would depend on the objectives of the instructing agent.
Hi Richard, really interesting! However I think all your 6 reasons still think of AGI as being an independent agent. What do you think of this https://www.fhi.ox.ac.uk/reframing/ by Drexler—AGI as a comprehensive set of services? To me this makes the problem much more tractable and better aligns with how we see things actually progressing.
If we avoid this dystopian near future of “superintelligent multi-level marketing” I hope the future will be more like this suggested by Steven Strogatz https://www.nytimes.com/2018/12/26/science/chess-artificial-intelligence.html which would leave the key remaining challenge being one of creating a mechanism for ensuring value alignment....
“But envisage a day, perhaps in the not too distant future, when AlphaZero has evolved into a more general problem-solving algorithm; call it AlphaInfinity. Like its ancestor, it would have supreme insight: it could come up with beautiful proofs, as elegant as the chess games that AlphaZero played against Stockfish. And each proof would reveal why a theorem was true; AlphaInfinity wouldn’t merely bludgeon you into accepting it with some ugly, difficult argument.
For human mathematicians and scientists, this day would mark the dawn of a new era of insight. But it may not last. As machines become ever faster, and humans stay put with their neurons running at sluggish millisecond time scales, another day will follow when we can no longer keep up. The dawn of human insight may quickly turn to dusk.
Suppose that deeper patterns exist to be discovered — in the ways genes are regulated or cancer progresses; in the orchestration of the immune system; in the dance of subatomic particles. And suppose that these patterns can be predicted, but only by an intelligence far superior to ours. If AlphaInfinity could identify and understand them, it would seem to us like an oracle.
We would sit at its feet and listen intently. We would not understand why the oracle was always right, but we could check its calculations and predictions against experiments and observations, and confirm its revelations. Science, that signal human endeavor, would reduce our role to that of spectators, gaping in wonder and confusion.
Maybe eventually our lack of insight would no longer bother us. After all, AlphaInfinity could cure all our diseases, solve all our scientific problems and make all our other intellectual trains run on time. We did pretty well without much insight for the first 300,000 years or so of our existence as Homo sapiens. And we’ll have no shortage of memory: we will recall with pride the golden era of human insight, this glorious interlude, a few thousand years long, between our uncomprehending past and our incomprehensible future.”
Hi Kit—Happy New Year!
Thanks for that, yes I hope a more digestible summary will be produced. I am not intending to be hostile at all, I am just very worried about the AI issue, just I see it as a different issue from that highlighted by the EA community, much more like that highlighted in the paper, hence my purpose in highlighting it.
I think humanity are not particularly generally intelligent, just they become programmed/conditioned to be relatively good at a number of tasks necessary to survive in their environment (eg. a baby chucked into the rainforest will not survive as long as all of the much less intelligent animals that live there). Indeed my worry is we are surprisingly stupid and manipulable as a broad group—as a species our driving motivators (fear, status) generally create the narrative in our cognitive consciousness, and our “blind spot” is the belief that we are much smarter than we are. in the US and the UK the political process has become paralysed as seemingly logical statements apparently talking to our conscious brain are actually playing to our deep subconscious motivators, creating a ridiculous tribalism far removed from any form of logic.
We perhaps “feel” intelligent as we create complex intellectual frameworks that explain things in detail, but this is really a process of “mapping the territory”. Its hollowness is shown in eg. Shogi, Chess and Go by Alpha Zero, since despite the many thousands of years of academic study poured into these subjects that has repeatedly mapped the territory this was blown aside by a self improving algorithm working out “what fits”. Maps in the real world might be good talking points but are simply nowhere near accurate enough at a human level of intelligibility.
As an investment banker I never had much interest in mapping the territory (despite being logical) but I was interested in “the best way to get from here to there avoiding the obstacles” (I did not care how as long as it works). And this is how life in generality (outside of academia) is—“how can i profit maximise doing x, without breaking any laws (better still if i find a clever way around the laws”. And with increasingly powerful self improving algorithms this ends up with the kind of dystopia shown in this video from Yuval Noah Harari and Tristan Harris—“supercomputers” (superintelligence) pointed at our brains. https://www.youtube.com/watch?v=v0sWeLZ8PXg
In all of this I know it is hard for EAs to properly engage—status is gained in any community (which is a powerful deep motivator) by largely agreeing with the norms of that community—and I know my views are far from normal in this community so status is gained by rejecting what I say. But as we share the same deep values—we want the world to be the best place it can be - (which is something very much other than making the most amount of money possible for already rich shareholders) - and I have huge belief in the potential and need for the EA community you will forgive me if I keep trying.
It appears unfortunate that nobody can be bothered to read the underlying paper, written by Eric Drexler, a senior Oxford Martin fellow which completely reframes the AI debate, to something far from paper clips and much closer to reality. If a human sat down at Chess, Go, and Shogi and simply by playing the game became far better than any other human in a couple of weeks we would all see this as superintelligent. That this achievement is so easily dismissed to me shows an a complete unwillingness to deal with reality as it is.
How about “rational altruists?” This to me is actually a better descriptor of the head and the heart than effective altruist, as a person could be really effective (on a QALY basis) without using the head at all—Live Aid was essentially emotionally driven, and drove a huge groundswell of support for tackling extreme poverty. The thing that sets effective altruism as currently named apart is the very high level of rational thinking that goes on in deciding what to do. Whether that is more or less effective than other approaches is probably an unhelpful starting point when it comes to outreach, as it can indeed sound arrogant, and ignores the fact that most people are emotionally driven in deciding how to give.
Wow this is a great post—thanks Katja!
My answer to the narrow question is that the idea has only recently emerged because of the recent emergence of social networks, allowing communities with a set of values outside the societal norm to emerge. As you describe it you thought deeply on your own about why people behaved the way they did, but on your own in a society with very particular values, effectively reinforced by marketing and group pressure you would have most likely simply conformed to your community norms over time (thinking about the meaning of life is not widely encouraged past university). Given that it requires an unusual level of “why” type thinking to focus on the issue this level of intellectual thinking was unlikely to achieve critical mass in any one particular physical location. That the movement is struggling to engage older mainstream groups is a demonstration of how deep conformance to social norms becomes, which only deepens I think the longer you are exposed to it, and makes the movement’s emergence online in a young intellectual community (which would be the first place to see critical mass) only to be expected.
Why the societal norm is to give so little and relatively so ineffectively despite the minimal “happiness” utility to an individual of the incremental dollar as incomes rise above a modest level in Western society (despite the prevalence of Christian values requiring a particular focus on caring for those in extreme poverty) is a deeper question. In my view it is probably due to slow steady evolutionary change of society, with no particular shock to the system to cause a widespread re-evaluation.
Until perhaps the 1960s rising incomes in developed economies were generally well aligned to rising happiness in a very real sense—cars, washing machines, TVs, central heating all improved peoples lives in very measurable ways. At this point it would have been very hard to argue in a coherent way that it was better to give to those you have never seen and knew nothing about (if you could meaningfully give to them at all) than to those you knew and loved close at hand.
The rising incomes that increased happiness created a strong almost pavlovian link between efficiency, profit and “a good thing”. With the introduction of the television powerful mass marketing became possible, increasingly playing at a sub-conscious level to our deep desires and motivations (fear, status, happiness) to create a need to consume more in order to allow profits to continue to grow as any capitalist organisation requires if it is to continue to flourish (which might otherwise have stalled). That, according to the World Happiness Report (page 3 http://www.earth.columbia.edu/sitefiles/file/Sachs%20Writing/2012/World%20Happiness%20Report.pdf) US GNP per capita has tripled since 1960 whilst happiness has remained relatively unchanged is an indication of the tenuous link that now exists in developed economies between growth and life improvement, but it is this key aspect that has allowed a “life improvement arbitrage” to be created that is now accurately observed and acted on by the effective altruism movement.
Now the concept has been created, and is to some extent obvious, it can be relatively easily understood by a very large group of people who with persuasion will wish to effectively invest in creating a better world. In selling the concepts to them though, and given the level of norming to societal benchmarks that has occurred I believe it will be necessary to use the same powerful marketing focussed on deep motivations to shift people’s behaviour as was necessary to allow the life improvement arbitrage to be created in the first place. There is no reason that this cannot be carried out with the same rigour of comparison as effective altruism would bring to any other activity—a commercial enterprise is rigorous in its assessment of the return on the marketing dollar, and there is no reason for the EA movement to be any different. There are many more things that EAs can examine now that the concept has been created.
I would probably suggest it ran out of Oxford as a separate project, as you can pick up the Halo effect. It wouldn’t sit well in a charity evaluator. Sits better alongside 80,000 hours—how do I make the most difference with my life—how do we make the most difference with our charity?
Fund a post for an EA to offer charity consultancy to charities to become more efficient in their particular mission.
The EA approach is quite transportable, and in demonstrating its effectiveness in tackling a wide variety of charity problems it allows real engagement with a broad mainstream charity population—charities are very set in their ways and new thinking is exactly what many charities need, as well as it would provide a receptive forum to the EA concepts and top charities—very powerful communication via tell, show and do
Having been an investment banker for many years working with top flight partners in Magic Circle firms in London my particular concern is that it can be very corrosive to long term morality, particularly as you rise to partner level. The moral choices you are often asked by clients to assist them in making in favour of financial return would I imagine affect most people. With the hours that such jobs require, minimising contact with an effective altruism community I would worry about a corporate lawyer losing their commitment to effective altruism as other moral guidance came to the fore.
This is absolutely so important. There is huge outreach potential for effective altruism into the mainstream. Millions of people would love to make the world a better place, and backing a movement full of vibrant young people doing exactly that using evidence and reason—well that is simply brilliant. Most older people will not want to get involved in the technical conversations—I have heard it distilled as “Go-Compare for charities” (and there involvement will therefore will not disrupt the heart of the EA community to respond to an oft quoted concern) but correctly positioned would be really interested in the sheer genuine, palpable, cut out the nonsense goodness of it—such a refreshing breath of fresh air in a cynical world.
The conclusion of the Oxford Martin Commission having studied the big problems of the world for many years were that at core of these problems is not something out there beyond us, as most of the big problems they identify having fairly clear solutions. The biggest problems are within us and systemic—short termism, selfishness and a narrow perspective.
The long term, cosmopolitan, and altruistic perspective of Effective Altruism is the perfect antidote for these problems, it just needs to go viral!
If EAs say deontological a bit less (particularly in talking to the mainstream) and be excited a bit more I believe that Effective Altruism can not only be one of the most exciting but one of the most important movements in the world.
Its great to see the number of meetups flourishing. As an “old geezer” I felt perfectly at home at the meetup organised in London last week, which was well attended and buzzing with interesting conversations. The link between the online community turning into a real community (which can reinforce giving behaviour etc) is really powerful in building strength and depth in effective altruism. If organisers of the events could drop quick meetup reports here we could build the knowledge base of what is working and what isn’t, what meetups could do etc it would probably also inspire people to start more.
Certainly i am inspired to start a slightly older group in London, with a link to and expertise from the core movement, which would provide two way learning—how to engage with the older mainstream, non-academic community, and start to normalise “making the most difference possible” as a normal, worthwhile thing to do in the mainstream community, and be inspired by an amazing community of young people who are doing exactly that. Thoughts/comments would be appreciated.
Just a thought—a huge number of people get involved in small charities for “self actualisation”/ego-enhancement and do so incredibly ineffectively, with the whole thing turning into an opportunity for self promotion which achieves very little at all—effective or not. Effective Altruists have a real opportunity to offer consulting services and provide guiding lights to doing more and doing it better on that particular cause, and in so doing gently introduce new (highly effective) causes to a broad cross section of the population. For example offering to provide xx hours of consultancy for $xx per hour (to be directly donated to your chosen cause) would:
assist your chosen cause;
help the assisted charity do what it does more effectively;
open the minds of the people involved in the particular charity to the concepts of “effective altruism” generally and your chosen cause in particular.
Thoughts/comments would be appreciated.
Absolutely. This is how all marketing campaigns work, and I am going to be starting on this shortly, it would be great to have you involved. In terms of the approach as Peter Singer said—Effective Altruism is a movement for the head and the heart. In outreach you need to use the head to capture the heart. I am working on the outreach strategy for Sustainable Human which has a following of 1.3 million on Facebook—all done by two guys over the last two years. Their top post is here—https://www.facebook.com/photo.php?fbid=10151541387192909&set=a.258016217908.138579.117609792908&type=1. Once you have that size of movement then you can normalise generous giving by creating a strong supporter network of local meet ups, most people once they are comfortable Give Well is honest will probably be content to just give to the top charities.
The point of departure has to be most people don’t think like Effective Altruists (you conduct mkt research to figure out how they do think and base your strategy on that research) but very many people want a happier fairer world and will be willing to pay generously to work towards achieving that, especially if they are part of a community that endorses and gives status to generous giving.
I think to outreach more broadly support mechanisms become crucial. For much of the founding community altruism has been a matter of philosophical choice. For others coming in from a broader background being part of a community that values altruism and happiness will help transition to more generous giving. Normalising generous effective giving will require support. Churches as an example are all about support, but it has made them very introverted. The balance is to make the support and community conditional on the giving and the outward looking.
I see no reason why EA cannot outreach broadly and stay true to its principles. Give Well I have heard described after a two minute explanation of what it did as Go Compare for charities (a price comparison site). Most people will be no more interested in the detail of how Give Well comes up with charities it does than how Go Compare works. To encourage greater giving a broad based genuine community needs to be created that makes greater giving normal. That there is an academic core is great, but that won’t be for many people. So I think the effect can remain thick, whilst the outreach can be broad.
A really superb book I have never seen highlighted by EAs is Human Purpose and Transhuman Potential:A Cosmic Vision for Our Future Evolution by Ted Chu, PhD. He has been the chief economist of GM amongst other things so he is very smart and well grounded, and it’s an amazing intellectual feat. I raise it as it covers all of the religious wisdom across all faiths within the book—I found it hugely thought provoking.