Artificial Intelligence: New Data Revolution
Artificial Intelligence: New Data Revolution
Cambridge University explores the exciting possibilities and dangers of rapidly developing artificial intelligence while questioning our preconceptions about AI.
Join the DZone community and get the full member experience.Join For Free
The new center at Cambridge University explores the exciting possibilities and dangers of rapidly developing artificial intelligence, while also questioning our preconceptions about what artificial intelligence is.
The purpose of this Center, which brings together many thinkers from different fields, is to investigate and predict the possibilities and challenges to be encountered as the development in the field of artificial intelligence accelerates, and to provide a more measurable and useful perspective on artificial intelligence. The Center for Existential Risk Research (CSER), who helped develop the new Center proposal, Dr. Seán Ó hÉigeartaigh points out that artificial intelligence has taken a large place on the CSER agenda: “This is partly because there has been so much progress in recent years; certain lower limits have been crossed, which shows that research has had a tremendous impact and is progressing very fast.
On the other hand, "Focusing only on disaster risk has constrained us in terms of the scope of this field, given that there is a lot to deal with in artificial intelligence," he underlines this point they are aware of.
The center is envisioned as a center that will house experts from similar disciplines dealing with artificial intelligence and examines not only long-term, short and medium-term effects, taking into account not only risks but also opportunities and challenges.
Different Approach to AI
Although artificial intelligence has made headlines or is the subject of films that challenge some of our concerns, the work of the new Center offers a different style and more applicable perspectives.
While the scary stories about Artificial Intelligence give a chill to our liking, its current use is relatively limited. Dr. Ó hÉigeartaigh explains: "Frankly, we can say that the AIs we see in the world are limited in intelligence as they are exceptional good practices for only performing certain tasks, such as navigating the city, playing chess, or running a search engine. At the moment, we have or can solve common problems. Even a dog does not have intelligence with cognitive abilities, let alone humans."
This measured and multifaceted approach has made the Center not only aware of the serious problems brought up by new technology but also open and willing to the opportunities it offers.
Dr. Ó hÉigeartaigh says that it is still only biological beings that can perform actions such as learning, adapting, thinking, and doing many other things in the world. "But defending the view that intelligence only occurs in biology is proof of the thesis that at some point we will grasp intelligence enough to re-create it."
Although Dr. Ó hÉigeartaigh's main field of study is computational biology, he has conducted interdisciplinary programs for many years. In this way, while managing the Existential Risk Center, he makes use of multidisciplinary thinking that deals with the researched issues from a wide variety of perspectives.
Dr. Ó hÉigeartaigh points out that “the answer to these big questions for the future is not just in computer science or computational biology” and “to ponder these long-term and picture-wide questions requires expertise in the political, economic, legal, sociological, and even philosophical domain”
Artificial Intelligence Applications
Indeed, the need for a different perspective is not only part of the intellectual chemistry that produces original thoughts and opinions, but also a partial answer to the question of how new technology builds knowledge and how it attracts different expertise. "
As a scientist, most of the challenges we face are that we have to analyze massive amounts of data from a wide variety of sources and make sense of incredibly complex systems that are interconnected. It's a very difficult thing even for multiple teams.
The systems we are currently developing make sense of big data. For example, helping the analysis of millions of genomes to find the origin of cancer, analyzing many aspects of climate change, or trying to make solar energy, our energy grid, or smart homes more efficient.
If we discover how to apply artificial intelligence to the problems we face, we can solve these problems. We will also be able to contribute. "
Social, political, and cultural aspects of the accelerating technological change lie deep within the scientific issues. Dr. Ó hÉigeartaigh, in this sense, addresses short-term concerns by citing taxis or long-distance drivers who lost their jobs due to driverless vehicles.
But he says that this situation can also give people time to do other things, which is exactly why different areas should contribute to these discussions.
While there are dangers that need to be addressed, such as the fact that artificial intelligence will enable the development of versatile drones shortly, Dr. Ó hÉigeartaigh points out that there is no artificial equivalent of human intelligence.
Rather than the more constrained intelligence currently used in many technologies, many failed predictions of more general artificial intelligence have long been made in the past. Dr. "Some may argue that the current enthusiasm is also wrong," says Ó hÉigeartaigh. "We also see unprecedented amounts of investments in this area and exciting projects focusing on more general approaches to artificial intelligence. While there is only a fifty percent chance of these happening this century, there should be people thinking and working on this topic."
It also indicates another important point: Even if an entire Artificial Intelligence fails, the technological advances made in this area will still be very important and the social, cultural, and political effects of these developments will have to be considered, discussed, and thought through.
Different Types of Intelligence
Another issue covered by popular discourses and discussions about Artificial Intelligence is that we handle this issue in a very anthropocentric way; However, we need to take into account that there are different types of intelligence in the world.
Dr. Ó hÉigeartaigh suggests that we have an approach that puts both human and planet earth at the center, from human intelligence to the intelligence of ravens from the family of carnivores: “We should not limit ourselves to anthropocentric intelligence.
One of the first projects that we defined in the inception phase is "Types of Intelligence", which we have already started to hold preliminary meetings about. " Among them is Murray Shanahan, Professor of Neurology at Imperial College, a specialist in bonobo intelligence, mathematical logic, and machine learning. "All of these people are focusing on finding relatively new ideas for different types of intelligence skills.
Although it is very difficult to say exactly what intelligence is, it may be easier to say what intelligence is for and start from here."
Another issue is how this type of Artificial Intelligence will evolve. Evolutionary biology has evolved by trial and error, and according to Dr. Ó hÉigeartaigh, some organisms with higher error rates developed faster than other organisms with lower error tolerance. "When designing our algorithms and artificial intelligence, we programmers have a choice of how we want to do it. There is also an artificial intelligence learning class that we call evolutionary algorithms that allow some degree of use of the trial and error method."
There are reasons why you want to be open to the changes that occur, as well as reasons why we do not want it, he says, “because we may not get anything significant at the end, or we may have unintended consequences”.
Many different evolutionary factors play a role at this point. Revolutions in scientific fields lead to more brainpower, more doctorate support, and the allocation of larger resources to the field of artificial intelligence, accelerating the development in this field at an explosive level. "An example of this is the high success of Deep Learning in its early days. This has enabled more resources and many highly successful people have used this method," he says.
Similarly, it is reasonable to say that there are conceptual breakthroughs in Artificial Intelligence, but it is not possible to predict how long it will take to make these breakthroughs or how much they will accelerate the developments in the field. "What we cannot predict creates great uncertainty, so it is absurd to say that we are certain that we will have general AI in 2070 just because we have made so much progress," he comments.
But sooner or later revolutionary breakthroughs will be made, and in this context, there is a need for places like Leverhulme that encourage people to think originally and creatively about things that will have great social impact.
The center also serves as a base where visitors from academia and industry will be hosted and school workshops and conferences will be held. "On this occasion, we aim to create a community that focuses on encouraging future opinion leaders and research leaders to address the issues that affect us all in the long run.
I am confident that the wonderful young people who come to our summer schools and workshops will play a major role in forming the future industry and political leaders in this field."
Published at DZone with permission of Daniela Tilkanen . See the original article here.
Opinions expressed by DZone contributors are their own.