Artificial Intelligence Feed

June 12, 2012

Self-Driving Cars: End of the Human Driving Era

It's 2030 and less than 1% of the current 190 million drivers in the U.S. can still legally drive. The reason: all 51 states (Puerto Rico was granted statehood in 2022) have prohibited human drivers from taking direct control of passenger vehicles while on public roads unless they are specifically trained and have a compelling reason. Farfetched?

The Case for Autonomous Cars

There are pressing reasons to get artificially intelligent, self-driving cars to market as quickly as it is safely possible to do so. Yet, while the idea of fleets of Google carbots delivering pizzas automatically is mesmerizing, the real justification to accelerate availability of this technology is the potential for a dramatic reduction in injury and fatality rates.

In the United States, the National Highway Traffic Safety Administration (NHTSA) recognizes the impact that both alcohol use and distracted driving have on public safety. Recent data shows that:

However, statistics in the United States compare quite well to most of the world. The World Health Organization (WHO) lists traffic fatalities as the #10 cause of death worldwide, accounting for 1.2 million deaths in 2008.

With approximately 3 trillion miles logged each year in the United States, there is an average of 1 death per 88 million road miles. Although much more data is needed to accurately gauge the safety prospects of self-driving vehicles, the track record is very promising. Of the 250,000 miles logged by Google's autonomous cars so far, two accidents have been recorded and both were the fault of human operators. Two things are certain: self-driving vehicle software won't drink while driving, and it actually can keep its attention on the road while posting Facebook updates.

Industry analysts believe that this technology won't be ready for public consumption for 10 years, but signs point to a more rapid adoption. The attention that car manufacturers are investing indicates a desire to make this capability available quickly. Google project manager Anthony Levandowski states that their self-driving cars are already completing test courses faster than humans and he thinks that a better-than-human safety record will be shown much sooner than the next decade.

Industry Embraces Autonomous Technology

In 2008, General Motors stated that they would begin testing driverless cars by 2015, and that they could be on the road by 2018. More recently, General Motors chairman William Ford, Jr. talked about a connection revolution at the ITS 2011 conference in Orlando. Similarly, GM's Vice President for Global Research and Development Alan Taub promoted technologies such as vehicle-to-vehicle communications, but worried about the significant challenge of still having humans at the wheel. In 2011, Taub stated:

The technologies we’re developing will provide an added convenience by partially or even completely taking over the driving duties. The primary goal, though, is safety. Future generation safety systems will eliminate the crash altogether by interceding on behalf of drivers before they’re even aware of a hazardous situation.

The impact of intelligent vehicle systems can already be seen in declining fatality statistics in the U.S. - down nearly 58% from 1.73 (per 100 million) in 1995 to 1.14 in 2009.

Evolving Legal Status

Once legal hurdles, liability issues and public perception challenges are overcome, and autonomous vehicles cars begin driving with injury and fatality rates lower human-driven vehicles drivers, the liability burden will shift. Unless a compelling reason can be found, humans must yield to self-driving cars that have a better, proven track record. While Taub says that driving for humans is "fun", fun is no longer an option when lives are at stake and better alternatives exist.

The shift to self-driving cars is not something that has occurred overnight. Although this technology has received a lot of press in the past few years, we've actually been moving towards greater vehicle autonomy for decades with capabilities such as adaptive cruise control, electronic stability control, collision warning systems, and lane departure detection to name a few.

Self-driving technology will initially require a competent driver be able to take control at a moment's notice and many governments are racing to make changes in the law. These changes seem premature since the legal and liability issues presented by this type of augmentation should not be different than with today's intelligent technologies - the human driver is still ultimately responsible for the vehicle operation at all times.

However, the question of liability will become murky at the point when self-driving vehicles can fully assume control of the vehicle from start to finish, and not require that a human driver be capable of taking control. A second round of changes in laws will be required, and questions of legal liability must be decided. Although driving will be much safer overall, in any particular accident, an assumption of fault will fall on autonomous technology until proven otherwise. Manufacturers and the developers of such technologies cannot afford millions of dollars in settlements for each incident. We must eventually make decisions to offer legal protection to manufacturers and the developers of these technologies since the benefits to the public as a whole are clear.



October 21, 2010

Predicting Stock Market Behavior with Social Networks

Stock Market Intersection
iStockphoto / James Steidl

Using a self-organizing fuzzy neural network model, researchers were able to correlate stock market movement 3 days in advance with a nearly 90% success rate by analyzing mood from a statistical sampling of tweets from Twitter.

Two mood measurement tools were used in the model. OpinionFinder measured public sentiment with simple positive / negative values while a new tool created by the authors called Google-Profile of Mood States, measured mood along six dimensions.

The paper discounts the Efficient Market Hypothesis (EMH) which states that, on average, returns greater than the market average can't be obtained because prices reflect all information that is currently publicly available. However, the authors don't take into account that while there may be a short-term opportunity to take advantage of such tools, once publicly available, these tools will themselves simply provide new sources of information that will be built into to market prices, thus reducing the window of opportunity for using them to an advantage.

A number of ways to increase the accuracy of the authors' model can be imagined. For example, an improved model might give proportional weighting to the number of subscribers that an information source reaches to take in to account influence.


December 12, 2009

Mathematically Predicting the Arrival of Human-Level Artificial Intelligence

Michael Anissimov, Steve Rayhawk, Anna Salamon, Tom McCabe, and Rolf Nelson have released the beta version of an application created from a Singularity Institute for Artificial Intelligence research project. The Uncertain Future predicts when human-level Artificial Intelligence will arrive based upon a rigorous mathematical model.

The application allows users to provide their own estimates as to the likelihood of key parameters. To assist you in determining the values of these parameters, links to noted schools of thought are provided with corresponding values spanning the gamut from true naysayer to advocate. Do you believe that Artificial Intelligence is impossible in principle like John Searle, or are you as optimistic as Stephen Hawking?

Check out Michael's blog about the topic and then visit The Uncertain Future to see what your estimates predict. What were your results?


November 22, 2008

Silicon Valley Artificial Intelligence MeetUp

iStockphoto / Kronick

Monica Anderson will speak about Model Free Methods and Artificial Intuition at the Silicon Valley Artificial Intelligence MeetUp at the TechShop in Menlo Park, California at 12 noon Sunday, November 23, 2008. The talk is one hour, followed by an hour discussion about any AI related topic.

Attendance is free but limited to 70 people. The event will be video recorded and may be posted on the web later.

Please join the AI MeetUp group and RSVP on the web site below if you want to attend:

The calendar page is updated regularly.

October 20, 2008

Brain Science Is About To Fundamentally Change

After inventing the Palm Pilot, Jeff Hawkins focused his efforts on neuroscience. He describes his memory-prediction framework theory of the brain in his book On Intelligence.

Predicting Patterns

This theory describes the process of how the brain makes predictions of future events by matching sensory inputs to stored memory patterns. Inputs that are processed from the bottom-up interact with expectations from the top-down to generate predictions. When a particular level recognizes a pattern, a label is associated and forwarded to the next level in the hierarchy.

Jeff Hawkins was inspired by an issue of Scientific American dedicated to the brain. He saw that neuroscience lacked a comprehensive framework to describe the operation of the brain and embarked on an effort to build one. In this TED video, he describes his ideas and their implications on artificial intelligence and machine learning.

September 11, 2008

Win Two VIP Passes to The Singularity Summit 2008!


The Singularity Summit 2008 is a proud sponsor of The Singularity Summit 2008 being held October 25, 2008 at the Montgomery Theater in San Jose, California, USA. This event is an annual event held by the Singularity Institute for Artificial Intelligence that gathers the smartest people around to explore the biggest ideas of our time.


The Prize

To promote free subscriptions to RSS news and email updates, is giving away two (2) VIP passes to this event to one (1) winner. These passes provide exclusive benefits and access to The Singularity Summit 2008 event that are not available to standard ticket holders.


How to Enter

To enter, send a single email to the following address:

Include your email address as the subject line, and your name and contact phone number in the body of the email.


Contest Rules

Contest is void where prohibited by law.

One entry allowed per email address. One winner will be randomly drawn from the entries submitted and will win two (2) VIP guest tickets to The Singularity Summit 2008, to be held October 25, 2008 in San Jose California, USA. Tickets are for admission only and include access to the VIP reception on Friday October 24, 2008. Tickets do not include any other associated items, including but not limited to, transportation and lodging.

Entries must be submitted and received by 12:01am EDT, Monday, October 6, 2008. The contest entry deadline may be extended at the sole discretion of The winner will be announced on the website.

Contest has no cash value. No substitution will be made in the event of cancellation or alteration of The Singularity Summit 2008. This contest is open to new and existing RSS and email subscribers, excluding employees and family members of

September 08, 2008

Loebner Prize Chatbots

Artificial Fiction Brain
Source: Wikipedia Commons. Licensed under Creative Commons Attribution ShareAlike version 2.5.

The Loebner prize in Artificial Intelligence is the first formal implementation of a Turing test with a substantial prize. The ultimate winner of this contest will receive US $100,000 and each year, approximately US $3,000 is given to the creator of the computer program that converses in the most human-like manner.

The Turing test was first described in Alan Turing's 1950 article Computing Machinery and Intelligence. Essentially, human judges engage in written conversation with both human and computer partners through computer terminals. The judges do not know ahead of time which of their partners are human and which are computer programs emulating humans. The computer programs have a goal of convincing the human judges that they are in fact humans, not computer programs. The idea is that human judges are unable to tell the difference between another human and a computer program, the computer program must in fact be, to some degree, exercising intelligent thought.

Although the Turing test is respected as a benchmark of Artificial Intelligence, the Loebner prize itself is somewhat controversial - the acclaimed "father" of artificial intelligence Marvin Minsky believes it to be nothing but an unproductive publicity campaign. This year's contest will be held at Reading University in the United Kingdom on October 12, 2008. You can chat online with some of the six finalists in advance of the contest:

Previous winners of the Loebner prize have been:

Some other memorable chatbots of interest:

  • The classic Rogerian psychotherapist Eliza.
  • Ally

Although not a contestant, another interesting program to chat with is This program plays a game of 20 (or more) questions with you and is quite often correct at guessing what you are thinking. You can help train it to get better by playing with it.

August 29, 2008

The Singularity Summit 2008

The Singularity Summit 2008

The Singularity Institute for Artificial Intelligence has issued a press release with details of The Singularity Summit 2008: Opportunity, Risk, Leadership. The event will be held October 25, 2008 at the Montgomery Theater in San Jose, California. Previous summits have featured Nick Bostrom, Eric Drexler, Douglas Hofstadter, Ray Kurzweil, and Peter Thiel.

Keynote speakers include Ray Kurzweil, author of The Singularity is Near, and Justin Rattner, CTO of Intel. At the Intel Developer Forum on August 21, 2008, Rattner explained why he thinks the gap between humans and machines will close by 2050. "Rather than look back, we're going to look forward 40 years," said Rattner. "It's in that future where many people think that machine intelligence will surpass human intelligence."

Other featured speakers include:

  • Dr. Ben Goertzel, CEO of Novamente, director of research at SIAI
  • Dr. Marvin Minsky
  • Nova Spivack, CEO of Radar Networks, creator of
  • Dr. Vernor Vinge
  • Eliezer Yudkowsky

To register for The Singularity Summit 2008, click here. You can find a comprehensive list of other upcoming worldwide Singularity and Artificial Intelligence events here.

Suggested Posts

Win Two VIP Passes to The Singularity Summit 2008!
Singularity Institute for Artificial Intelligence
What is the Singularity?
The Singularity Effect
Upcoming Artificial Intelligence Events
Distributed Computing Projects and the Singularity

Chris K. Haley, Subscribe Get free RSS or email updates here.

August 02, 2008

Upcoming Singularity and Artificial Intelligence Events of Interest

Update 6/17/2012: Hundreds of worldwide Artificial Intelligence, Robotics and Singularity conferences and events are now being tracked here:

Here are some events through the beginning of 2009 that are of interest to the Singularity, Artificial Intelligence, and Robotics communities:

Date Location Event
September 1-3, 2008 Palma de Mallorca, Spain Artificial Intelligence and Soft Computing (ASC-08)
September 1-3, 2008 Tokyo, Japan Eighth International Conference on Intelligent Virtual Agents (IVA 2008)
September 3-4, 2008 Hollywood, California Virtual Worlds Hollywood
September 4-6, 2008 Bulgaria 13th International Conference on Artificial Intelligence: Methodology, Systems, Applications (AIMSA-08)
September 28 - October 1, 2008 Technische Universität, Dresden, Germany 11th European Conference on Logics in Artificial Intelligence (JELIA-2008)
October 2-4, 2008 Syros, Greece 5th Hellenic Conference on Artiifical Intelligence (SETN-08)
October 25, 2008 Montgomery Theater, 271 S. Market St, San Jose, CA 95113, USA Singularity Summit 2008
October 26-30, 2008 Salvador, Bahia, Brazil 19th Brazilian Symposium on Artificial Intelligence (SBIA-08)
10th Brazilian Symposium on Artificial Neural Networks (SBRN-08)
Brazilian Symposium on Intelligent Robotics (JRI-08)
November 3-5, 2008 Dayton, Ohio, USA 20th IEEE International Conference on Tools with Artificial Intelligence (ICTAI-08)
March 6-9, 2009 Arlington, Virginia Second Conference on Artificial General Intelligence (AGI-09)

March 13, 2008

The Singularity Effect

The Singularity Effect
© / Konstantin Inozemtsev

The A.I. Effect describes a human cognitive bias to discount improvements made in the science of Artificial Intelligence. Problems that in the past that were seen as extremely difficult, or intractable, are now seen in retrospect as having obvious solutions which no longer need to be described in terms of artificial intelligence.

Similarly, I believe that a "Singularity Effect" describes the discounting of advances in other technology areas such robotics, genetics, nanotechnology, etc. Consider how some of these technologies would have appeared to an observer from even 50 years ago:

  • A camera which can detect and focus on faces, wait for people to smile before taking a picture and (soon) associate individual faces with names for indexing and future searching.
  • A neckband that translates silent vocalizations into speech. Although this device does not directly read human thoughts, two people using this device coupled with wireless capabilities would essentially appear to be telepathic - certainly to observers from decades past.
  • A computer operating system which achieves continuous speech recognition at normal speech rates and with high accuracy. It continually improves its accuracy by training itself on the speaker.
  • A neuroheadset which gives a computer a limited ability to read a game player's thoughts and emotions.
  • A tool accessible from anywhere for free which can generally search the most important areas of human knowledge and translate it to and from the most widely used languages, all within seconds.

Welcome to the Singularity!