Comment nous avons appris à faire confiance à l’IA


C2 Montreal is just an incredible event. The three day business conference is all encompassing, seamless, sexy, inspiring, shocking, and full of simmering controversies — and it’s expensive, and it ended last Friday.

Between 22–24 May some of the most brilliant minds in commerce arrived in Montreal to confabulate on their life’s work, the future, artificial intelligence and the collision of ideas at Montreal’s Arsenal space. Thankfully, the whole shebang is spread over three days, making it possible to make the whole affair feel substantial. And that’s part of what makes C2 so great — the sheer abundance of ideas, intriguing people to meet, and guest speakers from world renowned companies.
There were workshops that left us with knowledge we didn’t have before. Like the one with Dr. Carl Marci, a neuroscientist who specializes in how personal bias interferes with making accurate judgment calls. In just over 30 minutes, Dr. Marci proved that none of us is as smart as we think we are. It wasn’t even close. It was a humbling experience, and a little embarrassing.
The main events were held in the big top — a state of the art space with stadium seating, rejuvenating AC, a round stage where guests could circle like sharks, and a ceiling projection system that made us feel like we were at a Devo concert.

Here are some of the highlights:

On the subject of quantum computing, we were graced with a thoughtful talk by Dr. Robert Sutor from IBM. The qubits were flying as he rolled out one of the possible futures awaiting mankind: people working together, governments and corporations, unified under the banner of AI and exponential growth. Sutor was the first of many optimists.
Jessica Lauretti of RYOT Studio told us how virtual reality (VR) and augmented reality (AR) is finally where it needs to be to tell powerful stories that resonate with consumers. Maybe. Laurreti went on to discuss the Orlando shooting, which happened to have occurred on her first day at RYOT. She described getting her team together to figure out how they could produce a powerful 360 story around the tragedy. Something that would go viral. Apparently, the AR film performed very well. But her “audience-first” approach left us feeling cold. Since when does the audience know what it wants, anyway? Especially in an age when we’re all wrangling with algorithms designed to trap us and reinforce our biases 24/7?

If a storytelling company wants to impact real change and help avoid more tragedies like Orlando within the filter bubble era, we may want to look more closely at how some technology contributes to the radicalization of thought.

Buzzfeed sent the all-knowing Craig Silverman who showed us how machines have learned that humanity loves a good gross-out, and that the algorithms have fractured the way we construct and apply trust.
David Jones from John Deere explained how their magazine, The Furrow, has only mentioned the brand within its pages about half a dozen times since its inception in 1950. Take note, marketers; let the stories do the selling, not the deals.
Did we mention the children? Sabarish Gnanamoorthy, a very well-spoken 14 year-old VR and AR developer whose resumé makes us look like bums, told us that in the future, everyone will have their very own personalized, augmented personal reality to rule over. “Everyone’s reality will be different”, he said. The crowd ate it up. RIP, shared human experiences.
Stephanie Curallo, formerly at Apple, IBM and now COO of cloud content management service BOX gave a solid talk on what is necessary to transform a legacy workplace into a digital one. She spoke of changing business processes using Netflix as an example of a company that fundamentally pivoted 3 times in less than 20 years — from a product company, to a service company, to now creating their own content. She insisted on the idea of employees acting like entrepreneurs and leveraging the data to make sure the best idea wins. But before we even go there, shouldn’t we agree on what “best idea” means? Is it the one that generates the most impressions or the one that connects to our core human values and brings the most value to people’s lives?
Who else? Oh yeah, Skylar Tibbits from MIT blew our mind with game-changing, sci-fi-like advancements in 4D printing. Imagine, a sneaker that grows in water! Or when Eric Gutengeim of Cellink grew a human ear live on stage. The sheer breadth of guests reveals just how much is crammed into C2.
We were even treated to a fireside chat with human rights activist, cybersecurity expert and folkhero Chelsea Manning. Definitely a high point. Manning was alert with intelligence and easy to listen to as she waxed on about how each of us can be changing agents in the digital world. She spoke of the shocking militarization of police in the United States and reminded us of the dangers of human bias and AI in the age of mass surveillance. Ironically, her talk was followed by a conference by Facebook AI Research, presented by the Government of Québec. Later the same day she delivered a Masterclass: ethical tech in the age of mass surveillance. Strangely, this event was far from packed. Of all the radical talks, this one seemed pertinent to our everyday. But hey, it was 5 à 7 time on a hot and sunny Montréal afternoon — the digital constitution conversation didn’t stand a chance.
Snoop dog and Martha Stewart
Another favourite guest was an outlier named Bentley Meeker. An artist, lighting designer and CEO, who creates unique lighting environments for everything from concerts, to Burning Man, to galas, to Michelle Obama’s 45th birthday party. His talk was about humanity’s relationship with light, and how it’s in trouble. Let’s just say he’s not a fan of LED. He told us the only good light burns, and was the perfect segue to the brightest C2 star of them all.
Snoop Dogg. He took the stage late, obviously — the man’s a pro — and the big top exploded with fan’s cheers and whoops. Way more applause than any of the scientists got. Snoop was accompanied by his business partner, the always affable Ted Chung, along with their moderator, VICE’s Suroosh Alvi. They presented the duo’s cannabis brand to the Canadian market, talked weed and creativity, legalization in Canada, and with the help of the audience, a little about what justice might look like for the thousands of people serving prison sentences because they sold a little green. Also, we can’t not mention that Alvi was, in the words of Snoop himself, “high as a motherfucker.” Maybe take separate cars next time, guys.

Welcome to the 4th industrial revolution!

Most of our time at C2 was spent trying to make sense of what we were learning about artificial intelligence.
Montréal is no stranger to AI, having positioned itself as a major hub, attracting big brains from around the world at breakneck speeds due partly to the unprecedented support of government at each level. Many of the major players are in fact already here: Facebook, Google, Microsoft, with more on the way. The academic side is just as well represented with MILA (Montréal Institute for Learning Algorithms), and Element AI, a Montreal-based incubator that announced a Series A funding round of $102 million in 2017 and is now the go-to place for companies seeking to include AI solutions to their businesses. According to Yoshua Bengio — Canada’s resident machine learning guru — at least 100 researchers are exploring deep learning at the University of Montréal and about 50 others are doing similar work at McGill.
Speaking of Bengio — in a time when technology seems culture-encircling huge, it would have been nice to hear more from philosophers of his stature on how the technology will irreparably alter our souls? C2 should consider adding fiction writers, poets, musicians, artists and filmmakers who have something profound to say about tech in society. How are they coping with the looming 4th industrial revolution?
John Hamm
During the conference titled, ‘AI Gets Real’, we were treated to an insightful panel discussion with Navdeep Singh Bains (Canada’s Minister of Innovation Science and Economic Development), Jean-Francois Gagné (Element AI), and Sean Mullin (Brookfield Institute for Innovation + Entrepreneurship). Bains said we should get used to changing careers 15 times before retirement. Isn’t it sweet that he thinks we can retire some day? Seriously, it’s gonna be brutal. Seems like major disruption is just part of what it takes to build a true innovation economy. Just the way things have to be if we want what Bains called, “a slice of the pie.” It wasn’t comforting, to say the least.

This is it, the 4th industrial revolution, where the overarching attitude is: get ready, or else.

A theme was beginning to take shape. AI is advancing rapidly, but most of the public is still in the dark on how it will affect their daily lives. Yet there was a lot of talk about “trust”. Our government couldn’t stop mentioning it. Not just with our data, dollars and dopamine addictions, either. They want us to trust in the ideology. They want us to trust that artificial intelligence will pave the way to a world of borderless data and unhindered sustainable growth. If we can just learn to trust the machines (and the people who feed them), they’ll improve our lives, fix the environment and protect us from ourselves. AI will help us create more wealth, and we will all benefit from it, if we just let it.
And much like the industrial revolution before it, and the one before that, and the one before that, this brave new world will allow us to power new machines designed to reduce our dependability on human labour while also increasing production and profits. No industry is safe from what’s coming. But hey, it might be nice to see a few doctors on EI for once. Maybe we should introduce EI to AI? Their babies would be monsters, but boy are they good at chess.
Quebéc’s Premier, Philippe Couillard, also made a brief appearance at C2. He rallied the crowd (or tried to) about how bright the future looks for Canadians. Too bad he forgot to present us with any sort of plan for middle-class job creation in this utopia. Instead, we were left guessing as to what the plan really is. And with all these government figures spouting providence, one glaring question was conveniently omitted; how will AI be used for war?

The fight for the future

Paranoia and greed are at the heart of AI investments in many places, not to name names… ‘Merica. War, reach and profit are the driving force behind any revolution, and this one is likely no different. What happens when hardware becomes indestructible and artificial intelligence surpasses that of human beings? If people are to lose their jobs across all industries, wouldn’t that also include soldiers? What’s it look like when soldiers are on the dole? What happens when robots, powered by AI, see human life as nothing more than data packets, instead of horny, funny, flawed beings just trying to pay their bills on time?
Are we entering a global AI arms race? We followed Jean-Francois Gagné’s advice to AI skeptics and started digging. In an article published in the Verge, aerospace engineer Mike Griffin spoke about the Pentagon’s position on AI warfare. According to him, America’s adversaries, “understand very well the possible utility of machine learning. I think it’s time we did as well.”
Citizens do not have the luxury to sleep on this. Thousands of Google employees have already protested the role of the tech giant in the Pentagon’s “pathfinder’ AI program. Dozens of Google employees quit last May, refusing to be accomplices in America’s wars. Open letters like this one have been floating on the internet since at least 2015.

And yet, nobody at C2 spoke pragmatically of regulation. We need to act and demand a ban on Artificial intelligence for military use. Now.

Behold, the world’s first AI minister

One of the big surprises (or was it a late addition?) at C2 this year was the appearance of Omar bin Sultan Al Olama, the Minister of State for Artificial Intelligence of the United Arab Emirates — and the first AI minister in the world.
Are we entering a global AI arms race? We followed Jean-Francois Gagné’s advice to AI skeptics and started digging. In an article published in the Verge, aerospace engineer Mike Griffin spoke about the Pentagon’s position on AI warfare. According to him, America’s adversaries, “understand very well the possible utility of machine learning. I think it’s time we did as well.”

In AI we trust

If AI will shorten waiting lines at hospitals, make roads safer though driverless trucking, supercharge education, help us find true love (maybe even our keys) and lead more fulfilling lives, we’re all for it. Sign us up. But if there is even a small chance that this technology can put democracy at risk, thereby further fracturing our already teetering society, we need to have a real public conversation about AI.
This sort of tension is exactly why C2 is so special. Big ideas converge here, they pool and spill, elevating the vibrant discussions fuelled by savage cocktails, gourmet food trucks and beautiful people wearing mostly black. This wasn’t just a tech conference, a circus big top of ideas — this was drinks in the sun, sweaty handshakes and neck tans. It was electric boat rides along the canal, dance parties and photo-ops. And if you manage to stay relatively sober, you can learn a thing or two about where we’re all heading. Perspectives tussle and change shape, and if you’re lucky enough to get a ticket for C2 2019, you can take part too. It’s either that or be left behind. And nobody wants to miss a great party.
Tamy Emma Pepin is the chief executive officer and managing director of UPPL, a strategy, research and content firm built for the challenges of the digital world. Our team of thinkers and designers looks at systems and connects global trends across culture, tech, politics and media. Tamy is also the founder of ORB, a multimedia, multicultural and multidisciplinary space located in Montréal’s AI hub. ORB seeks to cultivate the sharing of knowledge across creative and scientific fields and encourage individuals to make meaningful connections through original programming, space and collaboration.
UPPL is in a paid partnership with C2MTL and had full freedom to write this piece.

Partager l'article

Dernières nouvelles