AI Zone Admin Forum Add your forum

NEWS: Chatbots.org survey on 3000 US and UK consumers shows it is time for chatbot integration in customer service!read more..

Mind|Construct moving forward
 
 

I started a new topic that I will use to keep everyone who is interested here, up to date on what’s going on right now with my project.

 

 
  [ # 1 ]

Dr. Hans de Bruin has officially joined the Mind|Construct science advisory board. He has been involved in the project for the last few years, acting as a scientific sparring partner. He will bring his very extensive experience in semantic knowledgebase systems to the Mind|Construct team.

http://www.mindconstruct.com/modules/news/article.php?storyid=13

 

 
  [ # 2 ]

Small world, apparently he works at my old college.

 

 
  [ # 3 ]
Don Patrick - Apr 3, 2014:

Small world, apparently he works at my old college.

Ah, so you attended the HZ in Vlissingen :D

 

 
  [ # 4 ]

Yup. Although I didn’t study ICT, the unlimited access to arrays of crashable computers and internet there allowed me to develop my programming skills. Even built my AI’s initial Javascript version there before college duties forced it into hiatus.

 

 
  [ # 5 ]

Just a little note on progress for the community here: I’ve had a meeting with my investor today and coming May 9 has now been flagged as the target-date for the official incorporation of the Mind|Construct organisation and subsequent shares emission needed to cater for the participation. 100K Euro will be made available before or on that date to cater for expenses linked to the incorporation and related things.

It seems we are moving forward grin

 

 
  [ # 6 ]

That sounds awesome, Hans Peter! Well done! smile

 

 
  [ # 7 ]

Congrats and good luck!

 

 
  [ # 8 ]

Sounds good Hans. Congratulations.

 

 
  [ # 9 ]

Thanks guys, I keep you all posted here grin

 

 
  [ # 10 ]

Today we launched our ‘new’ website. Although you might not spot the difference with our previous website incarnation, a lot has actually changed. We have build a new dedicated CMS from scratch, that is ultimately geared towards the specific information that we want to publish on our website.

http://mindconstruct.com/webpages/newsitem/18

 

 
  [ # 11 ]

Thanks for the yearly update, Hans. Good to see you again, however briefly. smile

 

 
  [ # 12 ]

HP, will your AI be integrated into your CMS, will you have any NLG functionality to demonstrate?

 

 
  [ # 13 ]

Dave, thanks for the re-welcome. The project did hit some bumps in the road over the last year, but we are in full swing again. More updates to come soon(ish).

Marcus, the AI will not be integrated into the CMS as that is not a target for the project. However, we are currently utilizing the same (code generation) technology for both the website CMS and the R&D platform we have build for the AI-system. More about that will be revealed soon wink Your question about NLG functionality will be answered in an upcoming news flash (probably somewhere next week), but don’t expect any public demonstrations soon, as we will reveal the system to the world when it is able to ‘seriously’ pass the Turing test (planned for Q1/2 2016). In the meantime there will be regular news updates to keep everyone posted on the progress.

 

 
  [ # 14 ]

Thank you for keeping us updated. I’m a little suprised that for the high aim this project has, you’re still planning to put it to a Turing Test. Does this not sidetrack from the path towards AGI, seeing as the focus in Turing Tests lies foremost in acting like a human?

 

 
  [ # 15 ]
Don Patrick - Apr 17, 2015:

Thank you for keeping us updated. I’m a little suprised that for the high aim this project has, you’re still planning to put it to a Turing Test. Does this not sidetrack from the path towards AGI, seeing as the focus in Turing Tests lies foremost in acting like a human?

Hey Don, yes I understand what you point at. But… although the Turing test ‘seems’ to point at ‘acting like a human’, I’m convinced that ‘acting like a human’ only really works if there is actual AGI ‘under the hood’. I don’t mean Turing test as in ‘gaming the definition’ like the Eugene Goodman event clearly did.

When I say ‘Turing test’, I mean an unlimited conversation between a human and our system, where the human will even be told up front that it is a machine he/she is communicating with and the communication should be so indistinguisable from a real human that we need to proof that it is actually a machine on the other end.

Funny enough, I had this scenario in mind for several years now (and actually posted about that on forums over the years), and now the new ‘Ex Machina’ movie is exactly about this idea of testing for ‘artificial consciousness’. I like to state up front (before everyone starts posting I’m wrong about the Turing test) that this is not exactly as Turing described the test, but this interpretation of the ‘goal’ of the Turing test seems to me a much better ‘test’ for Artificial General Intelligence.

 

 1 2 > 
1 of 2
 
  login or register to react