TAISU—Technical AI Safety Unconference
- [Open in Google Maps] [Open in local app]
- 14 May 2020, 10:00 am—17 May 2020, 7:00 pm
- 36 York Street, Blackpool, UK
- Contact: linda.linsefors@gmail.com
COVID-19 Update!
TAISU in it’s planed form is cancelled. But there will be a Web-TAISU around the same theme, and around the same time. I will make an announcement and probably open up for more applications when this thing is a bit more planed out.
Start: Thursday, May 14th, 10am
End: Sunday, May 17th, 7pm
Location: CEEALAR [formerly the EA Hotel], 36 York Street, Blackpool, UK
I ran this unconference in August last year. Since that event was a success, I decided to do another one.
This is an unconference around the question: “How do we design AI such that it is safe in the long run?” This includes things like alignment, corrigibility, transparency and also understanding humans and human value structures better. Questions regarding governance and coordination are not part of this particular event.
Because this is an unconference, it will become what you make of it. There will be an empty schedule which you, the participants, will fill up with talks, discussions and more.
You don’t have to stay the full length of the unconference. It is possible to join as many or as few days as you want to, and you are also welcome to stay longer at CEEALAR before or after the unconference.
Who can participate
To be able to have a high level of discussion during the unconference, I’ll require that all participants have some prior involvement with AI Safety. This year I’m also reserving at least half the spots for people who are currently working on AI Safety (as opposed to, have done so in the past, or are studying to do so in the future).
Applications
If this seems like something for you, fill out this application form. Application deadline is February 23rd. I will inform you who is accepted on February 28th at the latest (If you need to know sooner, let me know). If there is still room after that I will accept qualified participants on a first come first serve basis.
Update (Feb 24): TAISU is not full yet. I will now accept people on a first come first serve bases, for anyone who is qualified.
Price
£150 for the unconference, regardless of how many days you are attending.
If this is more than you can afford, let me know. If I think you belong at the unconference, I’ll give you as much discount as you need.
Food: All meals will be provided by CEEALAR, no extra charge.
Accommodation
You are responsible to book and pay for your own accommodation.
You can book a bed at EA Hotel there through their booking system. Choose the “Grant Application (working on EA projects)” option and let them know you are there for the TAISU. The stay at the hotel is pay what you want, but it will be appreciated if you pay £12/night, which is the cost price.
Alternatively you can get a room at any of the many nearby hotels.
- Announcing Web-TAISU, May 13-17 by (4 Apr 2020 11:48 UTC; 19 points)
- [AN #86]: Improving debate and factored cognition through human experiments by (12 Feb 2020 18:10 UTC; 15 points)
- Announcing Web-TAISU, May 13-17 by (EA Forum; 5 Apr 2020 22:26 UTC; 11 points)
- 's comment on rmoehn’s Shortform by (11 Feb 2020 6:21 UTC; 2 points)
COVID-19 Update!
TAISU in it’s planed form is cancelled. But there will be a Web-TAISU around the same theme, and around the same time. I will make an announcement and probably open up for more applications when this thing is a bit more planed out.
Official application deadline has now passed. Those of you who have applied to participate will soon get an email.
However, since TAISU is not full yet. I will now accept people on a first come first serve bases, for anyone who is qualified.
I wish i could make it, but i can’t. I’m not sure if i would meet the requirements either. I do have a plan for a group AI project that in part addresses the issue of AI safety. The project is integral to to fulfillment of the long term vision of the group so it cannot be evaluated separately from it, or from some of the other groups projects designed to support it. In order to explain it properly i would need to explain the concept of the group as a whole. I’m not trying to make it sound like it’s the best idea ever, or trying to seem clever. But I do believe that it might just be unique enough as a whole that it would not be a waste of your time to give it a look. We’re after the same thing with AI safety, I’m not after money or kudos, just trying to help. I made a google doc with and introduction to it and it’s projects. I can post it to my profile if that’s allowed.
Hi Jarson.
Due to the current pandemic TAISU will take a very different form than originally planed. I will organize some sort of online event on the same theme around the same time, but I don’t know much more yet. I don’t want to take on board more participants until I know what I’m organising. But ass soon as I know a bit more, I will do a new announcement and open up applications again. I expect this will happen with in a week or two.
Regarding your project, I’d bee happy to take a look at your google dock. Pleas share it.