January 2018 Newsletter

 |   |  Newsletters

Our 2017 fundraiser was a huge success, with 341 donors contributing a total of $2.5 million!

Some of the largest donations came from Ethereum inventor Vitalik Buterin, bitcoin investors Christian Calderon and Marius van Voorden, poker players Dan Smith and Tom and Martin Crowley (as part of a matching challenge), and the Berkeley Existential Risk Initiative. Thank you to everyone who contributed!

Research updates

General updates

News and links

Fundraising success!

 |   |  News

Our 2017 fundraiser is complete! We’ve had an incredible month, with, by far, our largest fundraiser success to date. More than 300 distinct donors gave just over $2.5M1, doubling our third fundraising target of $1.25M. Thank you!

 

 

Our largest donation came toward the very end of the fundraiser in the form of an Ethereum donation worth $763,970 from Vitalik Buterin, the inventor and co-founder of Ethereum. Vitalik’s donation represents the third-largest single contribution we’ve received to date, after a $1.25M grant disbursement from the Open Philanthropy Project in October, and a $1.01M Ethereum donation in May.

In our mid-fundraiser update, we noted that MIRI was included in a large Matching Challenge: In partnership with Raising For Effective Giving, professional poker players Dan Smith, Tom Crowley and Martin Crowley announced they would match all donations to MIRI and nine other organizations through the end of December. Donors helped get us to our matching cap of $300k within 2 weeks, resulting in a $300k match from Dan, Tom, and Martin (thanks guys!). Other big winners from the Matching Challenge, which raised $4.5m (match included) in less than 3 weeks, include GiveDirectly ($588k donated) and the Good Food Institute ($416k donated).

Other big donations we received in December included:

We also received substantial support from medium-sized donors: a total of $631,595 from the 42 donors who gave $5,000–$50,000 and a total of $113,556 from the 75 who gave $500–$5,000 (graph). We also are grateful to donors who leveraged their employers’ matching generosity, donating a combined amount of over $100,000 during December.

66% of funds donated during this fundraiser were in the form of cryptocurrency (mainly Bitcoin and Ethereum), including Vitalik, Marius, and Christian’s donations, along with Dan, Tom, and Martin’s matching contributions.

Overall, we’ve had an amazingly successful month and a remarkable year! I’m extremely grateful for all the support we’ve received, and excited about the opportunity this creates for us to grow our research team more quickly. For details on our growth plans, see our fundraiser post.


  1. The exact total might increase slightly over the coming weeks as we process donations initiated in December 2017 that arrive in January 2018. 

End-of-the-year matching challenge!

 |   |  News

Update 2017-12-27: We’ve blown past our 3rd and final target, and reached the matching cap of $300,000 for the Matching Challenge! Thanks so much to everyone who supported us!

All donations made before 23:59 PST on Dec 31st will continue to be counted towards our fundraiser total. The fundraiser total includes projected matching funds from the Challenge.


 
 

Professional poker players Martin Crowley, Tom Crowley, and Dan Smith, in partnership with Raising for Effective Giving, have just announced a $1 million Matching Challenge and included MIRI among the 10 organizations they are supporting!

Give to any of the organizations involved before noon (PST) on December 31 for your donation to be eligible for a dollar-for-dollar match, up to the $1 million limit!

The eligible organizations for matching are:

  • Animal welfare — Effective Altruism Funds’ animal welfare fund, The Good Food Institute
  • Global health and development — Against Malaria Foundation, Schistosomiasis Control Initiative, Helen Keller International’s vitamin A supplementation program, GiveDirectly
  • Global catastrophic risk — MIRI
  • Criminal justice reform — Brooklyn Community Bail Fund, Massachusetts Bail Fund, Just City Memphis

The Matching Challenge’s website lists two options for MIRI donors to get matched: (1) donating on 2017charitydrive.com, or (2) donating directly on MIRI’s website and sending the receipt to receiptsforcharity@gmail.com. We recommend option 2, particularly for US tax residents (because MIRI is a 501(c)(3) organization) and those looking for a wider array of payment methods.

 

In other news, we’ve hit our first fundraising target ($625,000)!

We’re also happy to announce that we’ve received a $368k bitcoin donation from Christian Calderon, a cryptocurrency enthusiast, and also a donation worth $59k from early bitcoin investor Marius van Voorden.

In total, so far, we’ve received donations valued $697,638 from 137 distinct donors, 76% of it in the form of cryptocurrency (48% if we exclude Christian’s donation). Thanks as well to Jacob Falkovich for his fundraiser/matching post whose opinion distribution curves plausibly raised over $27k for MIRI this week, including his match.

Our funding drive will be continuing through the end of December, along with the Matching Challenge. Current progress (updated live):

 



 

Correction December 17: I previously listed GiveWell as one of the eligible organizations for matching, which is not correct.

ML Living Library Opening

 |   |  News

The Machine Intelligence Research Institute is looking for a very specialized autodidact to keep us up to date on developments in machine learning—a “living library” of new results.

ML is a fast-moving and diverse field, making it a challenge for any group to stay updated on all the latest and greatest developments. To support our AI alignment research efforts, we want to hire someone to read every interesting-looking paper about AI and machine learning, and keep us abreast of noteworthy developments, including new techniques and insights.

We expect that this will sound like a very fun job to a lot of people! However, this role is important to us, and we need to be appropriately discerning—we do not recommend applying if you do not already have a proven ability in this or neighboring domains.

Our goal is to hire full-time, ideally for someone who would be capable of making a multi-year commitment—we intend to pay you to become an expert on the cutting edge of machine learning, and don’t want to make the human capital investment unless you’re interested in working with us long-term.

 

Read more »

A reply to Francois Chollet on intelligence explosion

 |   |  Analysis

This is a reply to Francois Chollet, the inventor of the Keras wrapper for the Tensorflow and Theano deep learning systems, on his essay “The impossibility of intelligence explosion.”

In response to critics of his essay, Chollet tweeted:

 

If you post an argument online, and the only opposition you get is braindead arguments and insults, does it confirm you were right? Or is it just self-selection of those who argue online?

And he earlier tweeted:

 

Don’t be overly attached to your views; some of them are probably incorrect. An intellectual superpower is the ability to consider every new idea as if it might be true, rather than merely checking whether it confirms/contradicts your current views.

Chollet’s essay seemed mostly on-point and kept to the object-level arguments. I am led to hope that Chollet is perhaps somebody who believes in abiding by the rules of a debate process, a fan of what I’d consider Civilization; and if his entry into this conversation has been met only with braindead arguments and insults, he deserves a better reply. I’ve tried here to walk through some of what I’d consider the standard arguments in this debate as they bear on Chollet’s statements.

As a meta-level point, I hope everyone agrees that an invalid argument for a true conclusion is still a bad argument. To arrive at the correct belief state we want to sum all the valid support, and only the valid support. To tally up that support, we need to have a notion of judging arguments on their own terms, based on their local structure and validity, and not excusing fallacies if they support a side we agree with for other reasons.

My reply to Chollet doesn’t try to carry the entire case for the intelligence explosion as such. I am only going to discuss my take on the validity of Chollet’s particular arguments. Even if the statement “an intelligence explosion is impossible” happens to be true, we still don’t want to accept any invalid arguments in favor of that conclusion.

Without further ado, here are my thoughts in response to Chollet.

Read more »

December 2017 Newsletter

 |   |  Newsletters

 

Our annual fundraiser is live. Discussed in the fundraiser post:

  • News  — What MIRI’s researchers have been working on lately, and more.
  • Goals — We plan to grow our research team 2x in 2018–2019. If we raise $850k this month, we think we can do that without dipping below a 1.5-year runway.
  • Actual goals — A bigger-picture outline of what we think is the likeliest sequence of events that could lead to good global outcomes.

Our funding drive will be running until December 31st.

 

Research updates

 

General updates

 

MIRI’s 2017 Fundraiser

 |   |  MIRI Strategy, News

Update 2017-12-27: We’ve blown past our 3rd and final target, and reached the matching cap of $300,000 for the $2 million Matching Challenge! Thanks so much to everyone who supported us!

All donations made before 23:59 PST on Dec 31st will continue to be counted towards our fundraiser total. The fundraiser total includes projected matching funds from the Challenge.


 
 

MIRI’s 2017 fundraiser is live through the end of December! Our progress so far (updated live):

 



 

MIRI is a research nonprofit based in Berkeley, California with a mission of ensuring that smarter-than-human AI technology has a positive impact on the world. You can learn more about our work at “Why AI Safety?” or via MIRI Executive Director Nate Soares’ Google talk on AI alignment.

In 2015, we discussed our interest in potentially branching out to explore multiple research programs simultaneously once we could support a larger team. Following recent changes to our overall picture of the strategic landscape, we’re now moving ahead on that goal and starting to explore new research directions while also continuing to push on our agent foundations agenda. For more on our new views, see “There’s No Fire Alarm for Artificial General Intelligence” and our 2017 strategic update. We plan to expand on our relevant strategic thinking more in the coming weeks.

Our expanded research focus means that our research team can potentially grow big, and grow fast. Our current goal is to hire around ten new research staff over the next two years, mostly software engineers. If we succeed, our point estimate is that our 2018 budget will be $2.8M and our 2019 budget will be $3.5M, up from roughly $1.9M in 2017.1

We’ve set our fundraiser targets by estimating how quickly we could grow while maintaining a 1.5-year runway, on the simplifying assumption that about 1/3 of the donations we receive between now and the beginning of 2019 will come during our current fundraiser.2

Hitting Target 1 ($625k) then lets us act on our growth plans in 2018 (but not in 2019); Target 2 ($850k) lets us act on our full two-year growth plan; and in the case where our hiring goes better than expected, Target 3 ($1.25M) would allow us to add new members to our team about twice as quickly, or pay higher salaries for new research staff as needed.

We discuss more details below, both in terms of our current organizational activities and how we see our work fitting into the larger strategy space.

 

Read more »


  1. Note that this $1.9M is significantly below the $2.1–2.5M we predicted for the year in April. Personnel costs are MIRI’s most significant expense, and higher research staff turnover in 2017 meant that we had fewer net additions to the team this year than we’d budgeted for. We went under budget by a relatively small margin in 2016, spending $1.73M versus a predicted $1.83M.

    Our 2018–2019 budget estimates are highly uncertain, with most of the uncertainty coming from substantial uncertainty about how quickly we’ll be able to take on new research staff. 

  2. This is roughly in line with our experience in previous years, when excluding expected grants and large surprise one-time donations. We’ve accounted for the former in our targets but not the latter, since we think it unwise to bank on unpredictable windfalls.

    Note that in previous years, we’ve set targets based on maintaining a 1-year runway. Given the increase in our size, I now think that a 1.5-year runway is more appropriate. 

Security Mindset and the Logistic Success Curve

 |   |  Analysis

Follow-up to:   Security Mindset and Ordinary Paranoia


 

(Two days later, Amber returns with another question.)

 

AMBER:  Uh, say, Coral. How important is security mindset when you’re building a whole new kind of system—say, one subject to potentially adverse optimization pressures, where you want it to have some sort of robustness property?

CORAL:  How novel is the system?

AMBER:  Very novel.

CORAL:  Novel enough that you’d have to invent your own new best practices instead of looking them up?

AMBER:  Right.

CORAL:  That’s serious business. If you’re building a very simple Internet-connected system, maybe a smart ordinary paranoid could look up how we usually guard against adversaries, use as much off-the-shelf software as possible that was checked over by real security professionals, and not do too horribly. But if you’re doing something qualitatively new and complicated that has to be robust against adverse optimization, well… mostly I’d think you were operating in almost impossibly dangerous territory, and I’d advise you to figure out what to do after your first try failed. But if you wanted to actually succeed, ordinary paranoia absolutely would not do it.

AMBER:  In other words, projects to build novel mission-critical systems ought to have advisors with the full security mindset, so that the advisor can say what the system builders really need to do to ensure security.

CORAL:  (laughs sadly)  No.

AMBER:  No?

Read more »