The Reverse Turing Test for AI

Turing Test
Google Duplex has been described as the world's most lifelike chatbot. At the Google IO event in May 2018, Google revealed this extension of the Google Assistant that allows it to carry out natural conversations by mimicking human voice. Duplex is still in development and will receive further testing during summer 2018.

The assistant can autonomously complete tasks such as calling to book an appointment, making a restaurant reservation, or calling the library to verify their hours. Duplex can complete most tasks autonomously, it can also recognize situations that it is unable to complete and then signal a human operator to finish the task.

Duplex speaks in a more natural voice and language by incorporating "speech disfluencies" such as filler words like "hmm" and "uh" and using common phrases such as "mhm" and "gotcha." It also is programed to use a more human-like intonation and response latency.

Does this sound like a wonderful advancement in AI and language processing? Perhaps, but it has also been met with some criticism.

Are you familiar with the Turing Test? Developed by Alan Turing in 1950, it is a test of a machine's ability to exhibit intelligent behavior equivalent to, or indistinguishable from, that of a human. For example, when communicating with a machine via speech or text, can the human tell that the other participant is a machine? If the human can't tell that the interaction is with a machine, the machine passes the Turing Test.

Should a machine have to tell you if it's a machine? After the Duplex announcement, people started posting concerns about the ethical and societal questions of this use of artificial intelligence.

Privacy - a real hot button issue right now - is another concern. Your conversations with Duplex are recorded in order for the virtual assistant to analyze and respond. Google later issued a statement saying, "We are designing this feature with disclosure built-in, and we’ll make sure the system is appropriately identified."

Another example of this came to me on an episode of Marketplace Tech with Molly Wood that discusses  Microsoft's purchase of a company called Semantic Machines which works on something called "conversational AI." That is their term for computers that sound and respond like humans.

This is meant to be used with digital assistants like Microsoft's Cortana, Apple's Siri, Amazon's Alexa or Bixby on Samsung. In a demo played on the podcast, the humans on the other end of the calls made by the AI assistant did not know they were talking to a computer.

Do we need a "Turing Test in Reverse?" Something that tells us that we are talking to a machine? In that case, a failed Turing test result is what we would want to tell us that we are dealing with a machine and not a human.

To really grasp the power of this kind of AI assistant, take a look/listen to this excerpt from the Google IO keynote where you hear Duplex make two appointments.  It is impressively scary.

Things like Google Duplex is not meant to replace humans but to carry out very specific tasks that Google calls "closed domains." It won't be your online therapist, but it will book a table at that restaurant or maybe not mind being on the phone for 22 minutes of "hold" to deal with motor vehicles.

The demo voice does not sound like a computer or Siri or most of the computer voices we have become accustomed to hearing. 

But is there an "uncanny valley" for machine voices as there is for humanoid robots and animation? That valley is where things get too close to human and we are in the "creepy treehouse in the uncanny valley." 

I imagine some businesses would be very excited about using these AI assistants to answer basic service, support and reservation calls. Would you be okay in knowing that when you call to make that dentist appointment that you will be talking to a computer? 

The research continues. Google Duplex uses a recurrent neural network (RNN) which is beyond my tech knowledge base, but this seems to be the way ahead for machine learning, language modeling and speech recognition.

Not having to spend a bunch of hours each week on the phone doing fairly simple tasks seems like a good thing. But if AI assistant HAL refuses to open the pod bay doors, I'm going to panic.

Will this technology be misused? Absolutely. That always happens, no matter how much testing we do. Should we move foraward with the research? Well, no one is asking for my approval, but I say yes.

Have You Noticed a Lot of Updates to User Agreements Lately?

lockedYou probably have received word via email or in apps lately about changes to company privacy and security agreements. Many companies are updating their privacy policy to make it "more clear and transparent." Why the sudden interest?

That was what a friend asked me recently. He surmised that it had "something to do with all the Facebook issues." That is partially correct. Having Mark Zuckerberg testify to the U.S. Senate and then to the European Parliament certainly put a spotlight on these issues.

But what really pushed companies was the EU's General Data Protection Regulation (GDPR) which went into effect this week. Since most websites are global, even if they don't think of themselves as being global, most big companies decided to adopt the GDPR standards for everyone, including their U.S. clients.

What I am seeing (yes, I read the fine print) is that they have added more detail about the information they collect, how they process that data, and how you can control your data. They may have updates on how they use cookies, for example, or how you can change who else gets to see your data. Some of these options have been around for awhile, but most users either didn't know about them or just didn't want to be bothered. For example, you have been able to block all cookies or third-party cookies or have them wiped when you close your browser for a long time. Did you ever change those settings?

These new changes seem to me to be a good and necessary next step. Add to the Facebook spotlight and GDPR the fact that Google's Chrome browser in its July 2018 version 68 release will mark all HTTP sites as “not secure.” Having the HTTPS  ("S" for secure) in that URL will become important. If your site appears to users as NOT SECURE, you can expect people to click away from it.

Blog Followers

I write regularly on five blog sites besides this one. It is always nice to see stats rise on the number of hits and visitors that come to the sites. Some blog platforms allow you to have "followers" - people who are notified when you post something new.

I have noticed something the past few months on two blogs I own that are hosted by Wordpress. There has been a marked increase in followers. That is a good thing, right? Well, yes but ALL of these new followers list an @outlook.com email address. I'm suspicious.

In early 2018, Outlook.com had a reported 400 million active users.That's a lot os users, but that number hasn't increased as much as Gmail's statistics.

But what might these new followers be plotting? Are they bots? Fake Russian accounts hoping to get into my blog and use it for nefarious purposes?

So far, nothing odd has happened concerning these new followers.

Has anyone else reading this found something similar happening with their blog or website?

 

PowerPoint Versus Narratives

Amazon founder and CEO Jeff Bezos writes an annual letter and in 2018 he repeated his rule that PowerPoint is banned in executive meetings. Bezos has also talked about this in public discussions. What does he prefer to those slide presentations? Narrative structure.

Narrative structure is something Bezos believes is more effective than slides. It is said that in Amazon meetings, you're not reading bullet points of text on a slide. Instead, Bezos says that everyone sits silently for about 30 minutes to read a "six-page memo that's narratively structured with real sentences, topic sentences, verbs and nouns." And then comes discussion.

You have probably heard the expression "death by PowerPoint." Slides (using PowerPoint, Keynote, Prezi, Haiku Deck or any other) can be deadly boring, but I still find presentation tools to be effective when used effectively.

Narratives, storytelling and discussion are great ways to learn and retain information. We know images also activate other areas of the brain and neuroscientists find that we recall things much better when when we see pictures of an object or topic than when we read text on a slide.

Text alone on slides is boring. It is bad presenting. But using slides with text, along with images, is one way to structure narrative and discussion. Every tool has its proper use and best applications. PowerPoint is no different.

Active Learning

Active learning is an approach that strives to involve students in the learning process more directly. That sounds so logical that I suspect some people would say "Isn't that what every class is doing?' It certainly is not a new idea, but it is not the norm in many courses and classrooms. 

I think the active learning approach was introduced by Reginald Revans as "action learning." Either term can describe an approach to have students do more than passively listening by being actively or experientially involved in the learning process.

logo

Frequently, this approach has students read, write, discuss, or be engaged in solving problems, and engaging in higher-order thinking tasks such as analysis, synthesis, and evaluation. A very simple definition might be having students doing things and thinking about the things they are doing.

I am doing a presentation this week that I titled "Predator and Prey: Active Learning Is Social Learning at the Active Learning Symposium at Rutgers University.

I base it on the premise that active learning is often social learning. The session will be primarily hands-on using a problem solving activity identifying animal species based on viewing skulls.

cat skullIt is a hands-on "active" presentation with people who have little or no background in osteology (the study of bones and skulls), but that is not what I am usually teaching when I do this activity.

I have used this activity with elementary school students, high school students, undergraduates and adults outside of a school setting.

I have usually used it in critical thinking classes, but the learners will also learn something about the skulls and species. When I use the activity to teach about osteology, it is an active way to involve the learners in critical thinking. Groups quite naturally are active and become social in the process. 

The action learning process typically addresses a real problem that is important, critical, and usually complex and involves a problem-solving set. The process promotes curiosity, inquiry, and reflection.

I'm not locked into labels and if someoen told me that my active learning activity was actaully experiential learning, or action learning, adventure learning, free-choice learning, cooperative learning, service-learning, or situated learning, I would say that is a good possibilty (though I know these terms are not strictly synonymous). My interest is in the learning, not the label.