Artificial Intelligence - Making it Right

The beginning of the year is extremely interesting for the artificial intelligence discussions and implementation of language technologies. After Google impressive step forward with the new features of Google Translate, Microsoft has put an emphasis on the language technologies making Cortana the heart of its operational system.

After the presentation of Windows 10 yesterday, I was hesitant about writing upon it, but I will keep it for the next week, as now it is important to take the helicopter view and go to the general concept of artificial intelligence. During 2014 we quoted a lot great personas and scientists discussing pros and cons of the artificial intelligence, and all of them agreed that the framework for its development should be set up with the technological advancements. This framework should surely have economic, social, technical and philosophical parameters. The same message has been expressed in the open letter issued by the Future of Life Institute[i] - the letter acknowledges “huge benefits” and “great potential” of the artificial intelligence techniques used as a tool and indicates the research directions and priorities including computer science framework, but also labor market economy and ethics[ii]. The letter is signed by many of the artificial intelligence researchers – scientists from Cambridge, MIT, Berkeley, etc., but also researchers from Google and Facebook.

The fact that the situation surrounding artificial intelligence advances is reaching its peak when it needs to be seriously taken into account on technical, but also economic and societal levels, is also confirmed by the fact that the discussions of artificial intelligence are gaining a major attention at the World Economic Forum in Davos. The briefings include questions analyzing the future of the labor market, like checking what jobs can be taken up by the smart machines in the upcoming years[iii]. Alison Gopnik explains brilliantly that the jobs including creativity aspect and changing values may be those that will still need human interaction, as understanding how creativity and emotions work is hard enough and people are still better at it than the machines. This sounds as an exciting perspective for all the humans working with creativity, arts, teaching.

Well, it seems that other researchers have come to the conclusion that creativity is not that complex to teach to the machines… A piece on the Financial Times Tech Blog quotes Demis Hassabis “chess wunderkind, games designer and one of Google’s artificial intelligence whizzkids”[iv] who thinks that creativity is not “necessarily this mysterious thing”[v] and “is something that we would hope our machines would have”[vi]. This principle lies in the use of neural network algorithms use – patterns and models similar to what is happening in human brain are “also a lot more efficient than doing complex probablistic calculations”[vii]. At the same time Demis Hassabis insists that it is extremely crucial to understand these patterns and “look inside” the neural networks to understand all possible deviations. Is not this a reference to the technological aspect of the artificial intelligence development and implantation framework discussed in the open letter quoted above? It surely is!

The take away from these discussions is that we are still on the way to discover and develop artificial intelligence – it is a powerful tool the humans have been able to create that is capable of great advancements. For getting the maximum of its use, the humanity should consider now all the aspects of its implementation framework. And technology is only one of them.

Should 2015 be a year when we will all work on the boundaries and rules for making the best possible of the artificial intelligence techniques we now have at our hands? At AnRCloud we are all in for making it happen so.

 

[i] Research Priorities for Robust and Beneficial Artificial Intelligence: an Open Letter by Future of Life Institute, available online at http://futureoflife.org/misc/open_letter, accessed on January 21, 2015

[ii] Research priorities for robust and benecial articial intelligence by Future of Life Institute, available online at http://futureoflife.org/static/data/documents/research_priorities.pdf, accessed on January 22, 2015

[iii] Issue Briefing: Artificial Intelligence, World Economic Forum, available online at http://www.weforum.org/events/world-economic-forum-annual-meeting-2015/player?p=1&pi=1&a=63931, accessed on January 21, 2015

[iv] Google’s DeepMind on the future of artificial creativity by Sally Davies on Financial Times Tech Blog, January 16, 2015, available online at http://blogs.ft.com/tech-blog/2015/01/256352/?, accessed on January 18, 2015

[v] Google’s DeepMind on the future of artificial creativity by Sally Davies on Financial Times Tech Blog, January 16, 2015, available online at http://blogs.ft.com/tech-blog/2015/01/256352/?, accessed on January 18, 2015

[vi] Google’s DeepMind on the future of artificial creativity by Sally Davies on Financial Times Tech Blog, January 16, 2015, available online at http://blogs.ft.com/tech-blog/2015/01/256352/?, accessed on January 18, 2015

[vii] Google’s DeepMind on the future of artificial creativity by Sally Davies on Financial Times Tech Blog, January 16, 2015, available online at http://blogs.ft.com/tech-blog/2015/01/256352/?, accessed on January 18, 2015