Unboxing Google’s 7 new principles on Artificial Intelligence

How many times have you heard that Artificial Intelligence (AI) is humanity’s biggest threat? Some people think that Google brought us a step closer to a dark future when Duplex was announced last month, a new capability of Google’s digital Assistant that enables it to make phone calls on your behalf to book appointments with small businesses. You can see it in action here:

The root of the controversy lied on the fact that the Assistant successfully pretended to be a real human, never disclosing its true identity to the other side of the call. Many tech experts wondered if this is an ethical practice or if it’s necessary to hide the digital nature of the voice.

turing-test3

Google was also criticized last month by another sensitive topic: the company’s involvement in a Pentagon program that uses AI to interpret video imagery and could be used to improve the targeting of drone strikes. Thousands of employees signed a letter protesting the program and asking for change:

“We believe that Google should not be in the business of war. Therefore we ask that Project Maven be cancelled, and that Google draft, publicize and enforce a clear policy stating that neither Google nor its contractors will ever build warfare technology.”

A “clear policy” around AI is a bold ask because none of the big players have ever done it before, and for good reasons. It is such a new and powerful technology that it’s still unclear how many areas of our life will we dare to infuse with it, and it’s difficult to set rules around the unknown. Google Duplex is a good example of this, it’s a technological development that we would have considered “magical” 10 years ago, that today scares many people.

Regardless, Sundar Pichai not only complied with the request, but took it a step further by creating 7 principles that the company will promote and enforce as one of the industry drivers of AI. Here are some remarks on each of them:

1. Be socially beneficial

For years, we have dealt with comfortable boundaries, creating increasingly intelligent entities in very focused areas. AI is now getting the ability to switch between different domain areas in a transparent way for the user. For example, having an AI that knows your habits at home is very convenient, especially when your home appliances are connected to the same network. When that same AI also knows your habits outside home, like your favorite restaurants, your friends, your calendar, etc., its influence in your life can become scary. It’s precisely this convenience that is pushing us out of our comfort zone.

This principle is the most important one since it bows to “respect cultural, social, and legal norms”. It’s a broad principle, but it’s intended to ease that uncomfortable feeling by adapting AI to our times and letting it evolve at the same pace as our social conventions do.

2. Avoid creating or reinforcing unfair bias

AI can become racist if we allow it. A good example of this happened in March 2016, when Microsoft unveiled an AI with a Twitter interface and in less than a day people taught it the worst aspects of our humanity. AI learns by example, so ensuring that safeguards are in place to avoid this type of situations is critical. Our kids are going to grow in a world increasingly assisted by AI, so we need to educate the system before it’s exposed to internet trolls and other bad players.

3. Be built and tested for safety

This point goes hand in hand with the previous one. In fact, Microsoft’s response to the Tai fiasco was to take it down and admit an oversight on the type of scenarios that the AI was tested against. Safety should always be one of the first considerations when designing an AI.

4. Be accountable to people

The biggest criticism Google Duplex received was whether or not it was ethical to mimic a real human without letting other humans know. I’m glad that this principle just states that “technologies will be subject to appropriate human direction and control”, since it doesn’t discount the possibility of building human-like AIs in the future.

An AI that makes a phone call on our behalf must sound as human as possible, since it’s the best way of ensuring a smooth interaction with the person on the other side. Human-like AIs shall be designed with respect, patience and empathy in mind, but also with human monitoring and control capabilities.

5. Incorporate privacy design principles

When the convenience created by AI intersects with our personal feelings or private data, a new concern is revealed: our personal data can be used against us. Cambridge Analytica’s incident, where personal data was shared with unauthorized third parties, magnified the problem by jeopardizing user’s trust in technology.

Google didn’t use many words on this principle, probably because it’s the most difficult one to clarify without directly impacting their business model. However, it represents the biggest tech challenge of the decade, to find the balance between giving up your privacy and getting a reasonable benefit in return. Providing “appropriate transparency and control over the use of data” is the right mitigation, but it won’t make us less uncomfortable when an AI knows the most intimate details about our lives.

6. Uphold high standards of scientific excellence

An open approach to building the AI systems of tomorrow is the best way of keeping any company honest. Providing access to “educational materials, best practices, and research that enable more people to develop useful AI applications” is a great commitment because it can expose problems faster and help find solutions sooner.

7. Be made available for uses that accord with these principles

It’s not an accident that this last principle is the predecessor of a section that outlines the applications that Google promises not to pursue. Many people fear AI just because they imagine what could go wrong if a faulty and uncontrollable system had the capability to make judgement calls on human behavior.

Google promises not to build AIs that can cause physical harm, promote surveillance that “violates internationally accepted norms” or any type of law/human right violation.

These principles are an invaluable start and a commendable attempt to establish the rules that will drive us towards the future. But they are not enough, proper regulation should be there to protect consumers. We are on the verge of a long journey involving AI that will change us, and we need to push forward and ensure that the proper guidelines are in place. Self-regulation cannot be the only type of control we see on AI. We need to push the tech industry towards the highest standards, for our future’s sake.

You can read Google’s principles here.


Did you like this article? Subscribe to get new posts by email:

Enter your email address to follow this blog and receive notifications of new posts by email.

Join 72 other followers


Image via University of Canterbury

Advertisements

What I learned moving from building enterprise to consumer software

Last April I decided to take a big jump from building enterprise software to building consumer products. I am very grateful to have found a place that would allow me to learn the ropes of the consumer business without sacrificing any of the internal goals. This past year has been a great learning experience with big learnings and here are my key takeaways.

Enterprise vs Consumer? What’s the big deal? 

Building enterprise software is a different beast than building it for consumers. They share several core components such as requiring a secure, reliable infrastructure and following best software practices including sprint models. However, I see three key differences.

Continue reading

Fixing Facebook’s privacy problem

Facebook has been receiving criticism once again for how they handled users’ personal data. Here is a quick summary: in 2013, a 3rd party developer acquired large amounts of data from about 50 million users through an old platform capability (which was removed by Facebook itself one year later to prevent abuse); this data was then used to target US voters during the 2016 Presidential Election. The issue is complex in depth and it highlights a bigger underlying problem: users’ privacy expectations are not aligned with the commitment from most tech companies.

Zuckerberg said in a recent interview with Wired, “early on […] we had this very idealistic vision around how data portability would allow all these different new experiences, and I think the feedback that we’ve gotten from our community and from the world is that privacy and having the data locked down is more important to people.”

Regardless, Facebook never committed to fully lock down users’ data, and their business model was in fact built around the value that data can have for advertisers through interest relevance and demographic targeting. Google and Facebook accounted for 73% of all US digital ad revenue in the second quarter of FY18, up from 63% two years before.

I can nonetheless relate to that idealistic vision between privacy and technology. The more information the Google Assistant knows about the music I like, the better it can personalize my listening experience. Richer actions become available too, like allowing me to control the Nest thermostat or the lights by voice. At the end of the day, I’m trusting Google with my music taste and the devices installed in my house, and I get the benefit of convenience in return.

Continue reading

4 lessons I learned losing money on Bitcoin

I looked at the “Buy Bitcoin” button and paused, was I ready to do it? had I read enough articles explaining what is blockchain? 2017 had just closed after an all-time high for cryptocurrencies, and according to many enthusiasts, it was just the beginning. I felt like I was missing out, so I pushed the button and sat back. I felt confident, but in reality, I had no idea what I was doing.

I passively consumed news about Bitcoin for years, but I never went deep enough to properly understand the technology behind it and its potential. Even though I followed the ultimate rule of “investing only what you can afford losing”, the truth is that I only began to comprehend blockchain technology after I already got my feet wet. I started losing money shortly after my first order completed, these are the 4 lessons I learned since then.

pexels-photo-315785

1. A big Bitcoin dive can drag the rest of the crypto market with it

There is so much speculation around cryptocurrencies and so many people investing in them without having a clue, that a moment of panic can snowball into a sudden market crash. A Bitcoin crash can affect many investors’ confidence in other cryptocurrencies (or altcoins), dragging their price down as well.

Many altcoins are variants of Bitcoin with small code differences, making their prices change practically in parallel to Bitcoin’s.

Continue reading

Amazon Go: A.I.’s grim face?

I have been waiting since college on RFID’s failed promise to deliver a walk-away checkout experience, and Amazon finally made it possible. After reading my co-blog writer’s experience in the Amazon Go store I had to check it out for myself and was excited for it. All my friend’s pictures were of long lines, but thankfully I am a morning person and there was no line when I got there. My goal was to pretend I had no idea what it was or how it worked. My experience overall was good, with the exception of the on-boarding process. I was greeted with a condescending “oh, you don’t have the app?” and was asked to stay aside. My T-Mobile reception was very poor so it took me a bit to get started. Once I downloaded the app and signed into my Amazon account everything was smooth. Mission accomplished! In this post, I’m not going to talk about the actual store (Ivan did a great job already) but about the implications of the first tangible and successful AI automated store.

20180126_163329061_iOS.jpg

Exterior of the Amazon Go store

Automation has always been part of our history. Automation has helped us evolve into the society we have now. Such as, automating how we grow and crop food so we can have a good food supply, the industrial revolution to make things faster and cheaper, the assembly line to make them even faster and cheaper, and finally computers to automate processes and tasks. Now, AI is here and it will automate all of our productivity.

Continue reading

I experienced the future of retail: Amazon Go

The craziest thing I’ve seen is someone who came in dressed in a Pikachu costume“, said an Amazon employee while she handed me a promotional bag with the Amazon Go logo on one side and the text ‘good food fast’ on the other.

I arrived at the new store in downtown Seattle around 7:20 pm and was surprised to see the line of people still reached the end of the block. It had been a cold day in Seattle but that didn’t discourage the hundreds of people who came to see the ‘magical’ store on day 0. I didn’t use the term ‘magical’ lightly here: the experience was truly unique and it felt too good to be true. Amazon Go is probably the store with more sensors on the planet right now, and it is intimidating:

The ceiling of the Amazon Go store

Each of those boxes on the ceiling are cameras connected to deep learning algorithms that analyze every move you make: which aisle you walk through, what items you grab to read and then return to the shelf, what items you put in your pockets or bag… everything to ensure you only get charged for what you take home. But also, everything to ensure your shopping pattern is studied and well understood. Maybe not today, but it’s the inevitable next step and the ultimate dream for any retail store: to know what their customers like and the type of advertisements that will work best on them.

Continue reading

Nintendo Labo: Thinking outside the box (or with the box?)

When Nintendo released the Switch last year I was very surprised by what they had been able to achieve, take the gaming industry on a spin (again). Once again they proved that they can innovate in a crowded space with deep pocket rivals. They were able to achieve something fun, flexible and that meets our new lifestyle not by thinking of specs but thinking of use cases. They understand people still want to play but they don’t do it just in a living room, so they would meet them where they are by providing play flexibility (great article about that here). Now, with Labo they have done something I consider priceless: enable kids to imagine, play and dream by connecting both the physical world and the digital one.

I have to be honest, I did not buy the Switch right away and when I did I played it and then returned it. Sometimes there is a price for innovation. To me, the Switch has two big drawbacks. First it is the lack of games. I could care less for Zelda (yeah yeah hate me) and some of the other games are just “meh”. However, it was the release of Mario Odyssey that finally made me get it. I loved it, it was fun, I could play at home and take it with me. I bought my Switch just before my holiday trip and took it on the road with me. This meant playing with the Joy-Cons inserted to make a huge Gameboy. I’m a big guy and I am very jumpy and move around when playing. Towards the end of the trip my Switch started to break. My gameplay would stop every minute because they would get disconnected (guess I can’t be that excited while playing). Turned out that the price to pay for the hardware flexibility was ruggedness. So when it was time to return, I could exchange or return and decided for the former due to lack of games.

I thought that would be the end of my Switch journey but this week Nintendo announced Labo. Nintendo has always been great at thinking outside the box. Some of these product work (Wii, Amiibos) and some don’t (VirtualBoy, Wii U) and that is the price to pay to try new things. What amazes me is Nintendo’s relentless pursuit of not thinking about what is the next big technology push they can do, but how to enable new ways to bring playfulness into our lives.

Continue reading

Are you rude to your virtual assistant?

2017 has been the year of the smart speaker. Amazon’s Echo Dot and Google’s Home Mini are currently selling for around $30, which makes them a popular Christmas gift. Using an Artificial Intelligence (AI) has never been cheaper and it’s finally reaching critical mass.

Companies are investing on AI more than ever: natural language recognition still has to improve a lot, but the current algorithms are already impressive. My favorite example: it’s now possible to ask “how long would it take me to get to Starbucks on 15th Ave?” and get an accurate response with the right assumptions. What a time to be alive!

All of this progress comes with side effects: having to learn how to talk to a machine. Often, people start talking without the wake-up keyword, and sometimes they forget to check if the device is actually listening, getting confused when there is no response to their inquiry. Talking to a machine is not easy and usually, very unsatisfactory.

Perhaps that dissatisfaction is what makes us be less aware about our manners when addressing an AI. What would you think if someone interrupted you mid-sentence with a sudden “STOP”? What if someone kept giving you orders relentlessly, never pausing to thank you? That’s how most of us talk to AI’s like Alexa or Siri, never saying “please” or “thank you”.

Continue reading

Why is Net Neutrality important?

In a few days, on December 14th, the Federal Communications Commission (FCC) will vote to change how the internet is regulated in the United States. The impact on how regular users like you and me access information online could be massive, and here’s why.

Currently, there is policy around open communications, mandating that the treatment of traffic should be non-discriminatory: broadband providers cannot get in the way, they cannot censor or make deals that benefit certain types of content over others. For example, Comcast is not allowed today to treat Netflix differently than any other new video streaming startup.

The FCC Chairman, Ajit Pai (who worked as Associate General Counsel at Verizon), is proposing to vote to remove these rules in favor of a less regulated internet. Fewer regulations means, according to their main argument, that the amount of investments made by telecommunication companies will stop decreasing. However, the same data that the FCC used shows that investments have been flat at worse (or actually increased at best) since 2013.

The FCC also argues that if any corporation starts misbehaving, consumers can take their business elsewhere. Nonetheless, more than 50% of the US population only has access to one single internet provider, so they cannot even vote with their wallets. The only way to defend consumers rights will be through litigation and class action lawsuits.

Is everything bad in the new proposed plan? No, there is one important point that will be critical to detect future abuses: broadband providers are obligated to be transparent about their traffic practices; in other words, if Comcast made a deal to promote Netflix over any other streaming service, or if they started blocking or throttling certain sites, it would have to be public knowledge (even if that just means one more line in the fine print at the end of a contract).

Continue reading

What’s in my phone’s home screen?

2017 is almost over, so I wanted talk about the apps that have taken the most important space on my phone during this year, and whether or not I think they’ll still be there next year.

Let’s start with a screenshot of my home screen:

I place apps in my home screen based on the frequency in which I use them. I try to minimize the amount of times I have to go to other pages of the home screen, so these are truly the apps that keep me going. But are all of these apps equally important for my daily tech routine? Will they stay in such a prominent position next year? Let’s break them into categories.

Connecting with friends & family

Messages, WhatsApp, Facebook Messenger and Mail are absolutely critical to stay connected with family and friends, especially those in other countries. I’m convinced that I’ll keep these around since they are literally the first thing I check every morning.

Facebook, Instagram and Snapchat have been part of an interesting migration during 2017: most of my friends stopped posting on Facebook and became more active on apps where their posts have a 24 hour expiration date. So far, most of my friends are choosing Instagram, probably due to the fact that it has a classic profile of everlasting posts; Snapchat will have a hard time recovering after the aggressive takeover from Instagram, so I would not be surprised if Snapchat didn’t make it on my phone through the next year.

Continue reading