Wired on AI and sentient computer systems
An interesting article from Wired entitled “AI systems will prove useful long before they become self-aware”.
Text and image source: Wired
“Artificial Intelligence (AI) is an umbrella term that is used to describe computers, robots, and software that mimic and act as human intelligence,” writes Disha Bathija. [“Artificial Intelligence: What Is In Store For Us,” Gizmodo, 4 July 2014] Bathija makes an important point about AI being an umbrella term. Too many articles on the subject assume that AI always refers to Artificial General Intelligence (AGI) which involves the pursuit of creating a self-aware (i.e., sentient) computer system. But, as tech investor Daniel Darling (@DanielDarling) asserts, “Artificial intelligence will help us long before we have our first proper conversation with a robot.” [“The real opportunities in artificial intelligence,” BRW, 2 July 2014]
As I have written in the past, there are lots of skeptics who doubt a truly sentient AI system will ever be created. Even if one does emerge, it will likely utilize a new kind of intelligence (i.e., different than human intelligence). According to Jaron Lanier, a computer expert and AGI skeptic, we can’t expect to duplicate something we don’t yet understand. He told Maureen Dowd, “We’re still pretending that we’re inventing a brain when all we’ve come up with is a giant mash-up of real brains. We don’t yet understand how brains work, so we can’t build one.” [“Silicon Valley Sharknado,” The New York Times, 8 July 2014] That’s why Darling states, “Forget sci-fi. Advances in artificial intelligence (AI) are being increasingly born out of practicality.” He continues:
“As all objects around us come online and start reporting information, the world is confronting a data tsunami. There is no way we can capitalise on this incredible jump in access to real time information without computer intelligence to make sense of it all. Fellow valley investor Om Malik got it right when he remarked, ‘instead of waiting for AI’s Godot – a machine we can converse with – what we really need are ways to use machine intelligence to augment our ability to understand our increasingly data rich and complex environment’. Achieving that will dwarf the pace of technical innovation to date and be a truly transformative force in the world.”
“All developments are cumulative in nature. Speaking of Artificial Intelligence, we are already interacting with its virtual avatar in the form of computer video games. When you’re playing a video game especially as single player, in games like chess, you are competing against computerized intelligence. One may not have realized it, but you have lost to computerized intelligence more times than you’d like.”
Ray Kurzweil, one the leading proponents of the search for a sentient computer system, notes, “[IBM’s] Watson is not quite at human levels in its ability to understand human language (if it were, we would be at the Turing test level now), yet it was able to defeat the best humans [in the game show Jeopardy!]. This is because of the inherent speed and reliability of memory that computers have.” [“Kurzweil Responds: Don’t Underestimate the Singularity,” MIT Technology Review, 19 October 2011] Whether or not Kurzweil and his colleagues achieve their goal, he underscores the fact that computers have inherent characteristics that make even weak AI systems useful. Bathija adds:
“One of the many advantages of AI is that its decisions are based on facts rather than emotions. It does not have an inherent intuitive mind or ‘gut feelings’ that humans operate from. The intuitive thinking in AI is developed on the basis of past behaviors of its user and data collected from every command entered in the computer’s system. Over-time algorithms and data pattern recognition technologies pick up on the behavior trends and psychology of its user and the AI starts generating results based on that. There is very little scope of mistake when you have historical and proven data to back up all decisions. Not just accuracy, we are also looking at non-stop working hours without a dip in efficiency or breaks required for food and refreshment. Unlike humans, machines with Artificial Intelligence do not need any sleep, thus overcoming the inherent disadvantage of tiredness in humans. Then there is the easy and 100% accurate knowledge transfer that can take place between two AI devices as against between two human beings where a lot depends on communication and perception. That eliminates any scope for information being lost in translation. Corporations have a lot to gain by AI replacing a lot of human functions relating to handling of big data and taking decisions based on that.”
The following video from PBS Digital Studios claims it can tell you everything you need to know about artificial intelligence in less than 9 minutes. That claim might be a bit hyperbolic, but the video does pack a lot of information into a short period of time.
As Maria Konovalenko (@mkonovalenko) writes, “Any brief overview of AI will be necessarily incomplete.” [“Artificial Intelligence Is the Most Important Technology of the Future,” Institute for Ethics & Emerging Technologies, 30 July 2013] She adds, “The key applications of Artificial Intelligence are in any area that involves more data than humans can handle on our own, but which involves decisions simple enough that an AI can get somewhere with it.” That statement applies to a vast number of activities across a wide array of human endeavor. To underscore the usefulness of weak AI, Konovalenko reminds us that back in 2009 a computer named Adam “became the first robot todiscover new scientific knowledge, having to do with the genetics of yeast.” She explains:
“The robot, which consists of a small room filled with experimental equipment connected to a computer, came up with its’ own hypothesis and tested it. Though the context and the experiment were simple, this milestone points to a new world of robotic possibilities. This is where the intersection between AI and other transhumanist areas, such as life extension research, could become profound. Many experiments in life science and biochemistry require a great deal of trial and error. Certain experiments are already automated with robotics, but what about computers that formulate and test their own hypotheses? Making this feasible would require the computer to understand a great deal of common sense knowledge, as well as specialized knowledge about the subject area. Consider a robot scientist like Adam with the object-level knowledge of the Jeopardy!-winning Watson supercomputer. This could be built today in theory, but it will probably be a few years before anything like it is built in practice. Once it is, it’s difficult to say what the scientific returns could be, but they could be substantial. We’ll just have to build it and find out.”
At Enterra Solutions, we are using artificial intelligence to create solutions for businesses that include the ability of the system to generate and analyze hypotheses on its own. We’re not trying to create a sentient machine, we’ll leave that to companies with deeper pockets, but we are exploring how cognitive computing can help business leaders make better and faster decisions.
Cognitive computer systems can handle routine decisions so that decision makers can concentrate their valuable time and effort on decisions that truly need the human touch. Although concerns remain that intelligent computers will continue to put workers out on the street, we believe that computers working with (not in place of) humans creates the most effective, efficient, and profitable working environment.