Can A Human Teach AI Ethics?

Google AlphaZero AI is taking humans one step closer to the Technological Singularity.   What happens if artificial intelligence surpasses h...

Google AlphaZero AI is taking humans one step closer to the Technological Singularity.  

What happens if artificial intelligence surpasses human intelligence? Experts are ever discuss and debating about the potential risks and as well as the great opportunities presented by modern technology especially AI which is still advancing and that it may outpaces the cognitive capacity of human being. Smart robots have became part of our daily fa-fairs. AI already took over the chessboard —up-next is to take the boardroom for commanding humans. Google’s AlphaGo is no ordinary cyborg but the first artificial intelligence that teach itself how to dominate the chess game without any human intervention. Did you heard this? It is evidently to state here that the technological singularity is ever unfolding closer.


AlphaZero is more advanced, developed with a “superhuman technical capability” in mastering the chess game in just four hours. Essentially, the AI system learns the tricks by absorbing humans' knowledge include entire history of chess game in sixth fraction of a day — and then figured out how to tackle anyone or anything. AlphaZero programmed with the rules of the game (not the strategy) AlphaZero played 100 games against Stockfish, the world champion chess program ever known. AlphaZero won 25. The last 72 games resulted into a draw with AlphaZero recording no losses and Stockfish recording no wins. “Now we know who is our new lord ,” David Kramaley, the CEO of chess science stated on the website Chessable. “It will no doubt that AI revolutionized the chess game, but think about this in a great picture; this means the its algorithm could be applied outside chess. The algorithm could run cities, continents, universes.” Google’s DeepMind lab has been working on this same technology for years, beginning with AlphaGo,which learned how to play the Chinese board game.

AlphaGo AI stomped to victory for a fourth time against Go world champion Lee Sedol.  

Using "reinforcement learning technique", where AlphaZero plays itself over and over again starting from random play and without any human supervision or data, as according to Google's DeepMind researchers stated in their study. This allowed the system to improve and refine its digital brain, known as a neural network, as it continually learned from experience. This basically means that AlphaZero it's own teacher. The new version of AI is more powerful than AlphaGo because it is no longer constrained by the limits of human knowledge. This is a major achievement for AI, and the subfield of reinforcement learning in generally. When AI teaching itself, the system sprung beyond unimaginable expectations and exceeding human knowledge by an order of magnitude in a short time span, while also developing unconventional strategies and creative moves.

The time when humans can have a meaningful conversation with an AI has always seemed far off and the stuff of science fiction. But for Go players, that day is here.

The Challenge: No doubt that, AlphaZero in next 5 years it will protrude hefty impacts on the rest of the world! According to Nick Hynes, a grad student at MIT's Computer Science and Artificial Intelligence Laboratory (CSAIL), confirmed that these specialised tool will have an impact on our daily lives. So, the decisions regarding the responsible design of artificial intelligence (AI) are often made by engineers who receive little training in the complex ethical considerations at play within their designs. Universities have struggled to find effective ways to integrate these issues into curricula for technical students. At the World Economic Forum Global Future Councils on Artificial Intelligence and Robotics, that comprised of leading academics and experts in the field, has recommended the creation of actionable AI map, social inquiry and as well as AI discourse.


The Opportunity: The project under AI world map seeks to link all professors from around the world and offers them a platform to share, learn and modify their curriculum to include a focus on AI ethics. This open-source sharing database will enable instructors to teach future AI scientists to understand the potential consequences of their work, empower them to make the right decisions as they design and deploy new products, and strengthen the use of responsible AI for the benefit of those companies and governments in which they end up working with.

Social Impact: The creation of AI without consideration of ethics and values, social inclusion and human centered approach can have deleterious consequences to individuals, businesses, and entire social structures. As recently demonstrated by the very controversial, Cambridge Analytica,reflecting on the use of AI to sway election results: “He thrilled to the intellectual possibilities of it. He didn’t think of the consequences.” AI scientists must be taught at the earliest stages how to consider and address issues such as bias, privacy, transparency and accountability in the creation of AI App or bots.

---