Three concerns on the Theory of Change, and a suggestion:
1. ToC's tend to be neglected once written (hopefully not by y'all!) especially as they show a possible route to impact but not timings and "who all does what and by when" to achieve the shared goal by a particular date. (In this case, that could be the date you realistically expect bad or misaligned ASI to entrench.) Have you considered doing a Critical Path plan (aka "Critical Pathway") as NASA did for Apollo?
2. Another problem is that the last stages of your plan rely on a single channel, with no redundancy. A critical path plan, which allows more than one route to achieving the goal of ASI safety implementation, could be life-saving?!
3. AI development is moving very fast, and this ToC would seem to be (a) quite long in time and (b) numerous in stages, which means lots of potential delays and failure modes, when
If you were in dialogue with any of ....
Met research, DSTL, GCHQ, Insider Threat teams, Global Threat teams, UCL Early Warning research centre, Kings College academics/military, UNDRR, NASA, ISRO's AI safety committe under Prof Singvi, etc
.... and doing any of the following:
joint scoping, workshops, seminars, training, prioritisation and threat/risk work
.... you might achieve prevention, preparedness or response capacity much sooner?
I'm happy to discuss further by phone, or link you in to some of those orgs.
I looked for an abbreviation for "soft nationalisation" and couldn't find one.
SN or S-N would create multiple problems and confusions.
At first I liked StNN, as it can be pronounced "stun" and is better as a hashtag and search term than SN, but it is already taken, even within the field: spatial-temporal graph neural network (STNN)
So I suggest one of these two:
BASSN = Bad Artificial Superintelligence, Soft Nationalisation
BAPSN = Bad AI Prevention through Soft Nationalisation
NB It's iodide, not iodine (potassium iodide, KI) and it MUST be pharmaceutical grade, NOT photo grade. If you see someone dosing out Lugol's iodine to kids, politely take it off them and use it for wounds or sterilising fruit/veg.
If you don't have tablets, but do have pharmaceutical grade KI powder, you need a very accurate balance to weight the correct dose for children, something like this: https://www.amazon.co.uk/dp/B07RZR5HYF
Buy it now. In an emergency, deliveries may be problematic.
Some other resources, which might be useful in other continents:
>If you believe in this community, you should believe in its ability to make its own decisions.
and these:
>Diverse communities are typically much better at accurately analysing ...
>The EA community is notoriously homogenous, and the “average EA” is extremely easy to imagine: he is a white male in his twenties or thirties from
I'm all for transparency, but it's not clear to me that normal "democracy" if it means "equal voting from the current EA constituency" is likely to make an improvement: as with some primary voting systems, it may lead to an even more narrow leadership culture.
A diverse representative "council" might be better, or something like citizen's juries and people's assemblies. Holocracy and Sociocracy would be well worth looking at, along with systemic concensing for 'minor' or 'low consequence' or short term decisions.
Participatory Learning and Action, participatory budgeting and participatory video are also very useful.
"Scaling Studies" are a thing, now part of "Implementation Science".
The focus tends to be what makes pilot projects scale-able, and what interferes with. Politicians and funders get (understandably) irritated when pilots keep failing to scale - this was happening a lot in the 1990s, which gave rise to the first studies on scaling.
Implementation science looks more generally at what really works IRL / in the field - a lot is going on in this in Chicago and Global Health.
Thanks for choosing an important topic (maternal health) which can greatly affect infant and whole family health and prosperity.
Historically, the interventions which lead to a very helpful 2 year gap between pregnancies are called "birth spacing" or just "spacing". If others are interested in this, or stunting, or impacts on educational attainment/IQ, a lot is available.
USAID's work on the BASICS programme and its successors is especially important, and produced a lot of effectiveness outputs and lessons learned, available now through me, in books, and via Scholar searches like this one on Ghana.
Some categories to bear in mind:
- for behaviour change in parental or sexual practices, single interventions of a limited duration are rarely effective; multiple tactics over an extended period work better; good examples include USAID BASICS approach to increasing breast-feeding and vaccination in Madagascar, using women's drama groups, radio, artists, flags on clinics, positive deviants, income-generation etc all in the same district
- a properly participatory approach (PLA) and (ideally) community-led approach is more likely to work and sustain itself; this takes more initial negotiation, and experienced community workers, but it pays off handsomely.
- this is an area where expertise and field experience in specific IRL context do matter: there are so many things to get wrong, and people have been working on effectiveness, health economics and RTCs for decades. This does not mean EAs have no role, but talking to larger NGOs/sector funders and asking them to identify gaps can be a good strategy. If an NGO repeatedly gets USAID, GIZ and French funding and has been around in a country for 20 years, they have probably achieved some good impacts and know some unmet needs and opportunities. NGOs are often friendly about networking and referals. (Try the Hash House Harriers or English-speaking church or embassy events, ot stop by at the Netherlands Embassy, often helpful and pragmatic)
I'm happy to be called about this, and any global health/poverty issues.
This is perhaps even more important post SBF scandal, where putting out a large amount of positive information for the public to find is quite crucial.
This seems very focused on EA and policy in USA and DC, rather than EA as a whole.
One of the original themes of EA was global health and poverty - it would be nice if you could do a part 2 with a wider focus? After all, DC has plenty of global policy think tanks, IFPRI, USAID etc.
Great to see this.
Three concerns on the Theory of Change, and a suggestion:
1. ToC's tend to be neglected once written (hopefully not by y'all!) especially as they show a possible route to impact but not timings and "who all does what and by when" to achieve the shared goal by a particular date. (In this case, that could be the date you realistically expect bad or misaligned ASI to entrench.) Have you considered doing a Critical Path plan (aka "Critical Pathway") as NASA did for Apollo?
2. Another problem is that the last stages of your plan rely on a single channel, with no redundancy. A critical path plan, which allows more than one route to achieving the goal of ASI safety implementation, could be life-saving?!
3. AI development is moving very fast, and this ToC would seem to be (a) quite long in time and (b) numerous in stages, which means lots of potential delays and failure modes, when
If you were in dialogue with any of ....
.... and doing any of the following:
.... you might achieve prevention, preparedness or response capacity much sooner?
I'm happy to discuss further by phone, or link you in to some of those orgs.
I looked for an abbreviation for "soft nationalisation" and couldn't find one.
SN or S-N would create multiple problems and confusions.
At first I liked StNN, as it can be pronounced "stun" and is better as a hashtag and search term than SN, but it is already taken, even within the field: spatial-temporal graph neural network (STNN)
So I suggest one of these two:
BASSN = Bad Artificial Superintelligence, Soft Nationalisation
BAPSN = Bad AI Prevention through Soft Nationalisation
NB It's iodide, not iodine (potassium iodide, KI) and it MUST be pharmaceutical grade, NOT photo grade. If you see someone dosing out Lugol's iodine to kids, politely take it off them and use it for wounds or sterilising fruit/veg.
If you don't have tablets, but do have pharmaceutical grade KI powder, you need a very accurate balance to weight the correct dose for children, something like this:
https://www.amazon.co.uk/dp/B07RZR5HYF
Buy it now. In an emergency, deliveries may be problematic.
Some other resources, which might be useful in other continents:
https://www.gov.uk/government/publications/radiation-emergencies-information-for-the-public/what-to-do-in-a-radiation-emergency#iodine
https://www.ready.gov/radiation
This is old, but has some good content:
https://apps.dtic.mil/sti/tr/pdf/ADA328301.pdf
There's a contradiction between this:
>If you believe in this community, you should believe in its ability to make its own decisions.
and these:
>Diverse communities are typically much better at accurately analysing ...
>The EA community is notoriously homogenous, and the “average EA” is extremely easy to imagine: he is a white male in his twenties or thirties from
I'm all for transparency, but it's not clear to me that normal "democracy" if it means "equal voting from the current EA constituency" is likely to make an improvement: as with some primary voting systems, it may lead to an even more narrow leadership culture.
A diverse representative "council" might be better, or something like citizen's juries and people's assemblies. Holocracy and Sociocracy would be well worth looking at, along with systemic concensing for 'minor' or 'low consequence' or short term decisions.
Participatory Learning and Action, participatory budgeting and participatory video are also very useful.
"Scaling Studies" are a thing, now part of "Implementation Science".
The focus tends to be what makes pilot projects scale-able, and what interferes with. Politicians and funders get (understandably) irritated when pilots keep failing to scale - this was happening a lot in the 1990s, which gave rise to the first studies on scaling.
Implementation science looks more generally at what really works IRL / in the field - a lot is going on in this in Chicago and Global Health.
Thanks for choosing an important topic (maternal health) which can greatly affect infant and whole family health and prosperity.
Historically, the interventions which lead to a very helpful 2 year gap between pregnancies are called "birth spacing" or just "spacing". If others are interested in this, or stunting, or impacts on educational attainment/IQ, a lot is available.
USAID's work on the BASICS programme and its successors is especially important, and produced a lot of effectiveness outputs and lessons learned, available now through me, in books, and via Scholar searches like this one on Ghana.
Some categories to bear in mind:
- for behaviour change in parental or sexual practices, single interventions of a limited duration are rarely effective; multiple tactics over an extended period work better; good examples include USAID BASICS approach to increasing breast-feeding and vaccination in Madagascar, using women's drama groups, radio, artists, flags on clinics, positive deviants, income-generation etc all in the same district
- a properly participatory approach (PLA) and (ideally) community-led approach is more likely to work and sustain itself; this takes more initial negotiation, and experienced community workers, but it pays off handsomely.
- this is an area where expertise and field experience in specific IRL context do matter: there are so many things to get wrong, and people have been working on effectiveness, health economics and RTCs for decades. This does not mean EAs have no role, but talking to larger NGOs/sector funders and asking them to identify gaps can be a good strategy. If an NGO repeatedly gets USAID, GIZ and French funding and has been around in a country for 20 years, they have probably achieved some good impacts and know some unmet needs and opportunities. NGOs are often friendly about networking and referals. (Try the Hash House Harriers or English-speaking church or embassy events, ot stop by at the Netherlands Embassy, often helpful and pragmatic)
I'm happy to be called about this, and any global health/poverty issues.
This is perhaps even more important post SBF scandal, where putting out a large amount of positive information for the public to find is quite crucial.
If EAs don't talk to journalists they will miss out on one really important learning:
.... how to talk to journalists!
This seems very focused on EA and policy in USA and DC, rather than EA as a whole.
One of the original themes of EA was global health and poverty - it would be nice if you could do a part 2 with a wider focus? After all, DC has plenty of global policy think tanks, IFPRI, USAID etc.
That's brilliant - is the work being extended to other countries?