Firefox browser offers new measures to counter background cryptomining – The Block Crypto

New security preferences have been added to the latest release of Firefox’s Quantum browser, the firm has written in a blog post. The company has rolled out features which include blocking unauthorised cryptomining and third-party fingerprinting.

Cryptomining or cryptojacking uses a computer’s CPU to mine cryptocurrencies without the victim’s knowledge or consent. A mining module is installed in-browser, usually slowing down the device and in the worst-case scenario even damaging the equipment.

Fingerprinting, on the other hand, allows a third-party to build a digital fingerprint to track someone’s behaviour across the web.

Firefox tested these anti-tracking solutions in a pre-release beta testing last month, after announcing the company’s approach in August 2018 and committed to rolling out features to mitigate these risks.

Local News, PAFE gives $118,723 to local educators – Bonner County Daily Bee

Barbie Hunt, Southside Elementary, Pend Oreille Water Festival, $1,125. Fifth-grade students will engage in a watershed education trunk, interactive lessons and finish with an outdoor field trip providing the opportunity to learn about watershed first-hand.

Ashley Delucchi, Sagle and Kootenai elementaries, KIndergarten STEM Kit, $449. This innovative grant will give kindergartners hands-on lessons in science, technology, engineering and math, with problem solving using magnets, motion and chain reactions.

Nicole Huguenin, Farmin-Stidwell Elementary, Makerspace: Final Frontier, $5,500. Students will receive the tools and supplies needed to create a Makerspace, a place to practice design thinking: empathize, define the problem, ideate, prototype, and test.

Adam Noble, Sandpoint Middle School, Solar Car and Grand Prix, $2465. Middle school students will plan, design, construct and later race a car that will operate on solar power, demonstrating their understanding of measurement, angles and aerodynamics.

Jewel Shea, Washington Elementary, Cougar Weather Station, $352. Students will analyze the weather station data to extrapolate local weather patterns and the connection to local and global climate patterns, then report it to staff and fellow students.

Kelly Krystinak, Southside Elementary, Growing Minds with Gardening STEAM, $799, This grant funds an indoor garden with grow lights, heirloom seeds, soil and potting materials allowing for a visual and hands-on botany experience for elementary students.

Jacque Johnson, Southside Elementary, Swamp Rock Fiddle Camp, $2.031. Students will take part in a string teaching residency focused on acquiring enough skills to be able to solo perform a three-cord fiddle tune with guitar accompaniment.

Ann Dickinson, Washington Elementary, Design for Change, $3,880. In addition to promoting complex problem solving and community involvement, students involved with Design for Change learn to apply academic skills such as research, writing, and speaking in authentic, meaningful contexts.

Nayla Morton, Sandpoint High School, CTE Students Run Business, $4,000. Graphic design students will have the opportunity to have hands-on experience in working with the customer in creating a custom product and business students will obtain industry experience running and managing a spirit store, including marketing, web development, accounting, and customer service.

Heather Guthrie, Kaleidoscope Art Education, District Wide, $2,500. This program offers academically and developmentally based art lesson that address Idaho state learning standards for the visual arts and common core standards for teaching and learning.

Jeralyn Mire, Sandpoint High School, Whats Next, $1,750. SHS Students will have a variety of opportunities to explore and experience the options available for life after high school through career/technical field trips, tours, job and college fairs and college visits.

James Koehler, Buster School Bus Safety, District Wide, $2,373. Promoting school bus safety throughout the 2019/2020 school year, the Buster Program will be involved in multiple community events and two national campaigns, School Bus Safety Week and Love the Bus Month.

Jacki Crossingham, Sandpoint Middle School, Klondike the Last Adventure, $1,600. Living Voices will perform a historic presentation focusing on The Klondike Gold Rush-The Last Adventure, serving as a springboard to begin the seventh grade English Language Arts Survival Unit.

Trish Butler, Kootenai Elementary School, No Drama Discipline, $2,590. Kootenai Elementary Staff will participate in a book study and training on No Drama Discipline, The Whole-Brain Way to Calm Chaos and Nurture Your Childs Developing Mind.

Lynette Leonard, Southside Elementary School, Smartlab: Creativity Learning, $7,000. Southside students will explore S.T.E.A.M. through applied technology and project-based learning in the integrated learning environment of a Smartlab, carefully designed to foster development of higher order thinking skills, where everything works together to support hands-on learning.

Randy Wilhelm, Lake Pend Oreille High School, Beekeeping, $560. This grant will purchase the protective gear for students to wear in order to learn the science and craft of beekeeping and the many aspects involved.

Shannon Kerrigan, Sandpoint High School, Training Connection Leaders, $1,500. This grant will provide professional, research-based training for selected SHS juniors and seniors who are nominated by staff and students to become Connections Leaders.

Carolyn Whalen, Learning with Lucky, District Wide, $9,000. Each first-grade student in the district will start the school year with a stuffed animal lab puppy that will serve as part of their daily reading team, proven to increase reading fluency and proficiency.

William Love, Sandpoint High School, Video Production Upgrade, $5,792. The grant will purchase professional-grade video cameras and equipment for journalism students in the Cedar Post and Monticola classes at SHS.

Dana Stockman, Sandpoint High School, AcaDeca Sickness and Health, $4,150. This grant gives SHS Academic Decathlon Program the opportunity to purchase curriculum materials and supplies to support the SHS AcaDeca Team for the 2019-2020 topic of In Sickness and in Health: An Exploration of Illness and Wellness, as they prepare across all academic subjects.

Becky Forell, Sagle Elementary, Mystery Science, $1,000. This grant allows a continued subscription to Mystery Science, impacting all students in grades K-5 at Sagle Elementary with this engaging, interactive and innovative curriculum.

Megan Jordan, Washington Elementary School, Literacy for Little Learners, $1,263. Kindergarten students at Washington will have hands-on materials to build a strong language arts base with the focus on phonemic awareness, letter knowledge, listening comprehension and vocabulary.

Elizabeth Delessio, Farmin and Kootenai Elementary Schools, LPOSD Summer STEM Camp, $4245. Summer STEM Camp is designed to engage students with hands-on opportunities in order to help prevent the summer achievement gap. Open to incoming 1st-6th students of Kootenai and Farmin-Stidwell.

Kendall Lang, Sandpoint High School, Now and Next Summit, $1,550. This grant provides students and their families with information and resources to assist students as they make decisions during high school in order to plan for their future and their post-graduation goals.

Tom Albertson, Sandpoint High School, PLC Assessment to Improve, $5,000. SHS will partner with Solution Tree to bring a guest instructor to speak about assessment practices to guide and measure student learning as part of the Professional Learning Communities for all teachers.

Shantelle Gillis, Northside Elementary School, Arts and Tech Integration, $5,000. This grant allows Northside Elementary to continue to implement an integrated arts skills development and art history curriculum, providing arts integration through age-appropriate education and supplies.

Jeannie Hunter, Sandpoint High School, Broadway in Sandpoint, $4,996. This grant provides royalty and rental fees for a major Broadway Play, to support and stage the production, showcasing student performance, stagecraft and technical work in planning and implementation.

Krista Jones, Southside Elementary School, Bal-A-Vis-X Exercise, $1800. Funding of this grant will make it possible for two teachers to attend the BLVX Training this summer as well as purchase the equipment to bring this multi-faceted mind-body system to Southside Elementary.

Nayla Morton, Sandpoint High School, Budget Challenge Simulation, $1,250. This grant provides registration fees for Budget Challenge, a personal finance simulation geared toward high school students to teach money management skills.

Natassia Hamer, Washington Elementary School, Leading Learning, $5,620. This grant will make it possible for Washington Schools Leading Learning Team to attend the Professional Learning Communities at Work Institute to then bring PLC to Washington Elementary.

Debbie Smith, Sandpoint High School, Future Diplomats, $1450. This grant will support SHS student participation in the National High School Model United Nations Conference held annually in NYC, building active, educated, globally-minded citizens.

Chantel Casey, Sandpoint Middle School, Professional Learning Communities, $2,500. Three SMS 7th grade Language Arts Teachers will attend the Professional Learning Communities at Work Conference, focusing on the roles and importance of Professional Learning Communities in our schools.

Kelly Curtiss, Sandpoint Middle School, Idaho Battle of the Books, $1,000. This grant brings Idaho Battle of the Books (IBOB) to SMS, a statewide voluntary reading motivation and comprehension program, exposing students to a vast variety of quality literature.

Melinda Rossman, Home School Academy, Arts. Drama. Arts. Culture, $5,000. This grant will make it possible to have at least one dramatic production each year and to bring in guests for cultural celebrations and education.

Paul Gunter, Forrest Bird Charter Schools, Ovations Performing Arts Education, $2,500. The Ovations Outreach Program will bring performing arts education to students in LPOSD through academic-based performance workshops that address Idaho State and Common Core Standards.

Jamie Parnell, Northside Elementary School, Summer Library Hours, $1,328. Students at Northside will have access to their library over the summer months to promote summer reading and engagement in an effort to continue to build skills as a reader and minimize learning loss.

LeeAnn Kopsa, Kootenai Elementary School, Summer STEM and Reading, $4,853. Summer reading and hands-on STEM projects will be paired for exciting summer opportunities for children and parents of Kootenai Elementary.

Kelli Knowles, Kootenai Elementary School, KT Learning Landscape, $851. This grant will assist in the development of an outdoor learning classroom at Kootenai Elementary. This space will be a multi-use addition to campus and will be in collaboration with Kaniksu Land Trust.

Alice Griffith, Farmin-Stidwell Elementary School, Keyboarding for Success, $547. This online keyboarding program will teach students proficiency at keyboarding, utilizing the self-paced, individualized program, Keyboarding for Kids.

Jeannie Hunter, Sandpoint High School, I Can See My Makeup, $390. The purchase of new make-up mirrors will benefit the theatre department, as students will use the mirrors to prepare for multiple productions each year, working on stagecraft and proper make-up application.

Pam Webb, Sandpoint High School, APSI: Literature, $769. This grant will allow teacher training and information for the newly revised Advanced Placement course exams for Language and Literature.

Aaron Gordon, Sandpoint High School, Brass Instruments, $869. SHS will receive a new brass instrument, a Bach Flugelhorn, benefiting the music program and adding to a full instrumental ensemble.

Lori Padilla, Book Trust Readers, District Wide, $4,000. This grant will help second graders purchase Scholastic Books each month in a program that works to improve reading skills by assisting in the choosing, ordering and building of ones own personal library.

Kristi Harrison, Hope Elementary School, Books for All, $1,500. This grant will help fund the Book Trust Program partnership with Scholastic to provide money toward books for K-3 students each month of the school year.

Canada lacks laws to tackle life and death problems posed by AI: Experts – Electronic Products & Technology

The role of artificial intelligence in Netflix’s movie suggestions and Alexa’s voice commands is commonly understood, but less known is the shadowy role AI now plays in law enforcement, immigration assessment, military programs and other areas. Despite its status as a machine-learning innovation hub, Canada has yet to develop a regulatory regime to deal with issues of discrimination and accountability to which AI systems are prone, prompting calls for regulation – including from business leaders.

“We need the government, we need the regulation in Canada,” said Mahdi Amri, who heads AI services at Deloitte Canada.

The absence of an AI-specific legal framework undermines trust in the technology and, potentially, accountability among its providers, according to a report he co-authored.

“Basically there’s this idea that the machines will make all the decisions and the humans will have nothing to say, and we’ll be ruled by some obscure black box somewhere,” Amri said.

Automated decision-making based on data that predicts

Robot overlords remain firmly in the realm of science fiction, but AI is increasingly involved in decisions that have serious consequences for individuals. Since 2015, police departments in Vancouver, Edmonton, Saskatoon and London, Ont. have implemented or piloted predictive policing – automated decision-making based on data that predicts where a crime will occur or who will commit it.

The federal immigration and refugee system relies on algorithmically-driven decisions to help determine factors such as whether a marriage is genuine or someone should be designated as a “risk”, according to a Citizen Lab study, which found the practice threatens to violate human rights law. AI testing and deployment in Canada’s military prompted Canadian AI pioneers Geoffrey Hinton and Yoshua Bengio to warn about the dangers of robotic weapons and outsourcing lethal decisions to machines, and to call for an international agreement on their deployment.

“When you’re using any type of black box system, you don’t even know the standards that are embedded in the system or the types of data that may be used by the system that could be at risk of perpetuating bias,” said Rashida Richardson, director of policy research at New York University’s AI Now Institute.

Time to move from lofty guidelines to legal reform

She pointed to “horror cases,” including a predictive policing strategy in Chicago where the majority of people on a list of potential perpetrators were black men who had no arrests or shooting incidents to their name, “the same demographic that was targeted by over-policing and discriminatory police practices.” Richardson says it’s time to move from lofty guidelines to legal reform. A recent AI Now Institute report states federal governments should “oversee, audit, and monitor” the use of AI in fields like criminal justice, health care and education, as “internal governance structures at most technology companies are failing to ensure accountability for AI systems.”

Oversight should be divided up among agencies or groups of experts instead of hoisting it all onto a single AI regulatory body, given the unique challenges and regulations specific to each industry, the report says. In health care, AI is poised to upend the way doctors practice medicine as machine-learning systems can now analyze vast sets of anonymized patient data and images to identify health problems ranging from osteoporosis to lesions and signs of blindness.

Carolina Bessega, co-founder and chief scientific officer of Montreal-based Stradigi AI, says the regulatory void discourages businesses from using AI, holding back innovation and efficiency – particularly in hospitals and clinics, where the implications can be life or death.

“Right now it’s like a grey area, and everybody’s afraid making the decision of, ‘Okay, let’s use artificial intelligence to improve diagnosis, or let’s use artificial intelligence to help recommend a treatment for a patient,”’ Bessega said.

Critics say Canada lags on exploring AI regulation

She is calling for “very strong” regulations around treatment and diagnosis and for a professional to bear responsibility for any final decisions, not a software program. Critics say Canada lags behind the U.S. and the EU on exploring AI regulation. None has implemented a comprehensive legal framework, but Congress and the EU Commission have produced extensive reports on the issue.

“Critically, there is no legal framework in Canada to guide the use of these technologies or their intersection with foundational rights related to due process, administrative fairness, human rights, and justice system transparency,” states a March briefing by Citizen Lab, the Law Commission of Ontario and other bodies.

Divergent international standards, trade secrecy and algorithms’ constant “fluidity” pose obstacles to smooth regulation, says Miriam Buiten, junior professor of law and economics at the University of Mannheim. Canada was among the first states to develop an official AI research plan, unveiling a $125-million strategy in 2017. But its focus was largely scientific and commercial.

In December, Prime Minister Trudeau and French President Emmanuel Macron announced a joint task force to guide AI policy development with an eye to human rights. Minister of Innovation, Science and Economic Development Navdeep Bains told The Canadian Press in April a report was forthcoming “in the coming months.” Asked whether the government is open to legislation around AI transparency and accountability, he said: “I think we need to take a step back to determine what are the core guiding principles.

“We’ll be coming forward with those principles to establish our ability to move forward with regards to programming, with regards to legislative changes – and it’s not only going to be simply my department, it’s a whole government approach.”

Principles on responsible AI use

The Treasury Board of Canada has already laid out a 119-word set of principles on responsible AI use that stress transparency and proper training. The Department of Innovation, Science and Economic Development highlighted the Personal Information Protection and Electronic Documents Act, privacy legislation that applies broadly to commercial activities and allows a privacy commissioner to probe complaints.

“While AI may present some novel elements, it and other disruptive technologies are subject to existing laws and regulations that cover competition, intellectual property, privacy and security,” a department spokesperson said in an email.

As of April 1, 2020, government departments seeking to deploy an automated decision system must first conduct an “algorithmic impact assessment” and post the results online.

Is Amazon Too Big? – Knowledge@Wharton

In an era when legacy retailers such as Sears and Macy’s are scaling back or going bust, online behemoth Amazon continues to boom. The company is the second-largest retailer in the United States behind Walmart, and last year it became the second company in the world to reach $1 trillion in market capitalization. Perhaps more significantly, it’s also one of the world’s largest tech companies, with reams of data collected from an enormous customer base. Amazon has sold 100 million units of its voice assistant, Alexa, and an equal number of Prime subscriptions. But is Amazon too big?

Amazon’s runaway growth has prompted questions about whether it has become a massive monopoly that has unfairly edged out smaller competitors. That question has caught the attention of consumers and policymakers, many of whom have called for legislation to regulate the global goliath. The question, while intriguing, seems to be a moot point for founder and CEO Jeff Bezos, who reportedly said his company is “not too big to fail.”

“I predict one day Amazon will fail. Amazon will go bankrupt. If you look at large companies, their lifespans tend to be 30-plus years, not a hundred-plus years,” Bezos said, according to a report by CNBC.

The Knowledge@Wharton radio show on SiriusXM invited two experts to discuss the potential backlash to Amazon’s unbridled growth. Wharton marketing professor Barbara Kahn and Ryan Hamilton, marketing professor at Emory University, tackled the topic during a recent interview. (Listen to the podcast at the top of this page.) Following are key points from their conversation.

Regulation Isn’t the Same as Antitrust

The power amassed by Amazon and other tech giants hasn’t gone unnoticed by American politicians. Last year, President Donald Trump launched a Twitter attack against the company and The Washington Post, which is owned by Bezos. In March, Sen. Elizabeth Warren, D-Mass., unveiled a plan to break up Amazon and other companies making $25 billion or more in revenue. But the professors pointed out that subjecting big tech to antitrust laws, which take aim at monopolies that have pushed out competition, is different than enacting legislation to regulate that sector.

“Typically, when you think about antitrust, you think about whether the consumer is worse off. And Amazon has been so far pretty clean on that,” Kahn said, adding that Amazon hasn’t lowered product quality or raised prices. The company also appears to be transparent with its customers.

However, more than half of American households now subscribe to Amazon Prime, so Amazon controls a massive amount of online shopping data in an asymmetric way against the competition.

“There’s nothing that Amazon isn’t touching.” –Barbara Kahn

“I think there are a lot of different issues,” she said. “Historically, what people think about Amazon is that they have so far been true to their word and their consumer focus. But they are ruthless against the competition.”

According to Hamilton, the issues provoked by Amazon require an entirely new way of thinking about antitrust. Monopolies have long been about the ability to control price and profit. Amazon could get broken up into 20 different entities, but that wouldn’t solve the problems surrounding access to data. The issues with data management could be addressed through legislation, which Hamilton thinks is more likely than the government intervening to fracture the companies.

“People are talking about breaking up Amazon; people are talking about breaking up even Facebook, which is a free service. So, it’s no longer about just price; it’s about information, power. It’s about leverage, about control. I would be surprised if there’s kind of an appetite for that broadly, but opinion on this seems to be shifting rapidly,” he said.

Amazon Is Everywhere

Kahn offered a simple, succinct description of Amazon: “There’s nothing that Amazon isn’t touching.”

Most consumers are familiar with Amazon’s online shopping and streaming services, but the company really makes its money from the cloud-based Amazon Web Services (AWS) and Prime subscriptions, which generally cost $119 a year.

“They can offer a very good price and very high quality because they’re really not making much money on margin,” Kahn said. “That, to me, is part of the genius. I don’t think it’s thinking necessarily through a retail lens, it’s thinking through a very sophisticated business lens.”

Amazon is also buying real estate, specifically empty retail sites that can be used as fulfillment centers, and stepping on Google’s toes with a foray into targeted advertising.

“There was a headline [recently] where Jeff Bezos announced that Amazon is planning to go to the moon,” Hamilton said. “So, when we talk about everywhere and everything, you may not be able to escape Amazon by leaving the planet.”

Part of Amazon’s genius is closing the traditional “last mile” for the customer, Kahn noted. That means the burden is no longer on the customer to drive somewhere in order to research, comparison shop and purchase. With a click, the products arrive quickly at the doorstep.

“Amazon has just been ruthless at identifying customer pain points and eliminating them.” –Ryan Hamilton

Amazon continues to try to close the gap. In May, it announced a $10,000 incentive program for employees who want to start their own Amazon same-day delivery service. Bezos is also looking at replacing warehouse workers with robots to achieve greater speed and efficiency.

“Amazon has just been ruthless at identifying customer pain points and eliminating them,” Hamilton said. “When they first started off, it was supposed to be a cheaper way to get books and to find books that might be hard to find elsewhere. Then they just moved down the ladder: What’s the next customer pain point we can get rid of? Three days is too slow? All right, two days. Two days? No, one day. Soon, they’re going to get it to you before you even order it.”

The ‘Creepiness Factor’

Privacy issues are starting to crop up for Amazon, although the company hasn’t faced the same backlash as Facebook. Still, customers have expressed concern about Alexa listening in on their conversations. Hamilton calls it the “creepiness factor.”

“One of the ways that we can overcome this creepiness factor is when people see a clear upside, right? If they understand the benefits that they’re getting from more invasive technology, then people are kind of more accommodating to it,” he said. “I think that that’s where Amazon has avoided some of the problems that Facebook and Google and others have run into, where it seems really obvious to the consumer the benefits that they’re getting from Amazon. Amazon is ramping it up and faster and faster. It doesn’t breathe. Better and better suggestions on what you should buy next. I think as long as that’s transparently obvious to the consumers, there’s going to be less resistance.”

Kahn agreed, saying the transparency is what gives consumers more confidence in Amazon than, say, Facebook.

“They still haven’t violated consumer trust yet, except for privacy issues, but you haven’t seen them do anything inappropriate with that information — that we know of,” she said. “What’s interesting is I go around and give a lot of speeches on Amazon, and when I talk to consumers, they are so loyal to Amazon Prime, it’s shocking, you know? They can’t imagine their life without it.”

“I go around and give a lot of speeches on Amazon, and when I talk to consumers, they are so loyal to Amazon Prime, it’s shocking.” –Barbara Kahn

Always Pushing Forward

Both professors attribute Amazon’s outsized success to the innovative culture instilled by Bezos. While the company doesn’t have the best reputation for the way it treats employees, whatever he’s doing is working. Bezos is the richest man in the word, with a post-divorce net worth of $110 billion.

“I think that Amazon is changing so much of the way we think about business,” Hamilton said. “A standard problem that a maturing organization runs into is that it starts to get entrenched in the way that it’s doing things and then fails to miss the next opportunity. Amazon has so far not fallen into that trap at all. They are just constantly blowing things up and constantly looking for new ways to solve problems.”

But Hamilton pointed out that Amazon can’t continue to do everything. At some point, companies require a focus on core competencies. He’s not sure Amazon has that focus.

Kahn disagreed, saying Amazon may have gotten its start as a retailer, but its real focus is on data and an ever-expanding customer base.

“They have more and more interaction, so now it’s not just bodies but the frequency of interaction with those bodies, and suddenly you have an advertising model,” she said. “So, I think what they’re doing is creating a network and creating in-depth information about that network and different ways to get at that network, and that is what their focus is. Once you have that, you’ve got a business model. You can start selling all types of different services and content to that base.”

Don’t Discount Walmart

As big and powerful as Amazon has become, it has a formidable competitor: Walmart. Kahn said the discount retailer is right on Amazon’s heels. When Amazon began offering one-day delivery to Prime members in certain areas, Walmart announced free next-day delivery on its most popular items.

Walmart is better than Amazon on cost, operational excellence and logistics, Kahn said. And the retailer is using those proficiencies to its advantage. Even its revamped website is easier for shoppers to navigate than Amazon’s, which Kahn describes as “built for an engineer.”

“I think that Amazon is changing so much of the way we think about business.” –Ryan Hamilton

Hamilton isn’t quite as sold on Walmart, saying the retailer will not be able to beat Amazon at its own game. People will continue to shop in stores for certain things, so a physical presence will always be needed. But Walmart will have to chart a different course than Amazon.

“I think that Walmart is doing some smart things, but things are changing so rapidly. I know that sounds like kind of a commentary without any teeth, but things are changing so rapidly that what has worked in the past is not going to have any kind of salience going forward,” he said. “They’re leveraging the strengths that they have in terms of their physical presence, but they’re not going to be able to become the next Amazon. They’re going to have to become something different if they’re going to succeed.”

Kahn flatly disagreed.

“I do think Walmart can beat Amazon at its own game, not in terms of being everything to everyone, but in retail, they can. They’re still the world’s biggest retailer, and they compete head to head in a lot of the same categories. They’re growing very fast online,” she said. “Walmart understands in retail in its bones, whereas Amazon is kind of flirting with it. So, I think that you will see great things from Walmart. I’m very pro-Walmart. What they’re doing is very interesting.”

Edited Transcript of BIDU earnings conference call or presentation 17-May-19 1:15am GMT – Yahoo Finance

Towards the end of Q1, we began to see the impact of macro environment, compounded by a significant release of ad inventory into the market, causing CPM post Chinese New Year, to not rebound as pronounced as we had historically experienced. Although the Chinese government has announced many economic policies to bolster the economy, given the current macro conditions, tighter government scrutiny on content, cut backs from the VC community and so forth, we are taking a cautious view that online marketing in the near term will face a more challenging environment.

We are reviewing our resource requirement and improving our execution capabilities, which Herman will discuss in more detail later. While greater operational efficiency will be a focus for us, our priority will be to strengthen our mobile business and make the necessary investments to position Baidu for higher, long-term growth.

We are working on a CRM offering for our marketing customers, which we plan to release later this year that will feature better user targeting, customer engagement and sales force management. Coupled with our performance-based marketing services, our Platform-as-a-Service, will allow our marketing customers to move from user interest and intent targeting to live messaging or call to order taking in a seamless process.

Our CRM app will consolidate Baidu’s marketing services placement systems, including search, feed and Baidu Union, for easier campaign targeting and management. Baidu CRM will leverage Baidu’s massive user insight, our sales force management experience and technologies and our AI capabilities.

For example, our smart invitation feature provides personalized prompts to invite users to chat online with sales reps when Baidu AI matches the user interest with relevant customer offering. To better manage its sales force and better understand their customer calls, our customers can run a report to identify sales agents who were rude on a customer call and summarize comments made by customers during sales calls, leveraging Baidu’s advanced Natural Language Processing capabilities.

In March, Baidu’s mobile reach expanded to 1.1 billion monthly active devices, providing rich intent and interest-based user insights. Our focus on in-app services, expanding our ecosystem of content and services and optimizing user relationship management offer an exciting future for us to grow, to continuously improve our user experience, stickiness and mind share. Strengthening our in-app services and user relationship management are the foundation that allows us to go deeper into the marketing funnel to offer our customers CRM capabilities beyond our leading performance-based ad.

Beyond the 1.1 billion mobile devices, we hope to expand Baidu’s platform reach to DuerOS smart devices in homes and in autos and out-of-home digital screens which will enhance Baidu’s omni-marketing capabilities.

Our vision for AI-first Internet is to provide the best user experience to our users as they move across Baidu’s family of 20-plus apps as well as prepare our users to move from mobile-only to a cross-platform experience, seamlessly crisscrossing back and forth on mobile, in homes and in autos. We are quite excited about the opportunities ahead, for example, to offer cross-platform distribution capabilities to Baidu’s content providers and service app developers.

To keep pace with the fast-changing technology environment, Baidu has adopted a young leadership development program to incubate talent for middle management and above, to align with the market that we serve and provide an option for leaders to retire, pursue their personal interests and release their responsibilities.

Over the past year, excluding iQIYI, Baidu added 7 Vice Presidents, both newly promoted and new hires, to our strong executive team of over 20 members. Hailong Xiang, our Senior Vice President of Search business, has tendered his resignation which I have accepted. I’d like to thank Hailong for his 14 years of service at Baidu.

I’d also like to congratulate Dou Shen for his promotion to Senior Vice President, overseeing Baidu’s mobile business, which we have renamed from our search business. Dou previously served as Vice President of Baidu’s mobile products and has served in various other roles at Baidu since joining in 2012, including web search, display advertising and financial services group. Prior to Baidu, Dou worked at Microsoft and cofounded Buzzlabs. Dou holds a PhD in computer science from the Hong Kong University of Science and Technology.

Let’s begin our Q1 review with search plus feed. We continue to increase the scale and bolster the content and service offering in Baidu App to give our users unmatched native app-like experience for search and feed. Combining feed with search increases user stickiness. We are employing this strategy with Haokan and investing in Video Search.

DAU for Baidu App and Haokan are growing robustly. In March, Baidu App DAUs reached 174 million, up 28% year-over-year, and Haokan short video DAUs reached 22 million, up 768% year-over-year. Total feed time spent on Baidu App and short video apps grow robustly, up 83% year-over-year in March.

We continue to expand Baidu’s ecosystem of third-party content and services, which enable news feed and short videos as well as services and related information traditionally found only in apps, searchable on Baidu. Baijiahao, our feed content network, now hosts 2.1 million publisher accounts, up 89% year-over-year as more and more top publishers, creators and media companies are taking advantage of Baidu’s large search plus feed distribution scale.

Our Smart Mini Program, including its user base and developer network, continue to scale. MAUs of Baidu’s Smart Mini Program in March reached 181 million, up 23% sequentially. Service providers, large and small, are adopting Baidu’s Smart Mini Programs as we leverage our search capabilities. For example, Baidu’s Smart Mini Program has become the largest traffic channel for a leading social commerce company in China behind its mobile app. Notably, our AI-powered algorithms help male users on the Baidu platform find interesting content in the social commerce company’s Smart Mini Program, increasing a user group that has been historically underrepresented. Baidu’s Smart Mini Program allows users to search by the content within the Mini Program, employing our strong search capabilities.

Baidu’s Smart Mini Program is also helping users to find less frequented apps with long-tail content. For example, an online urban dictionary adopted Baidu’s Smart Mini Program and saw its user experience dramatically improve, with daily search traffic increasing by 27x and Baidu’s Smart Mini Program becoming its most significant source of traffic.

In addition, we are strengthening our vertical offerings, such as health care and online literature, filtering out poor and questionable content with Baidu AI. We are adding mini games in the Baidu App with more vertical offerings to come.

Through the Chinese New Year Gala campaign, we educated users that Baidu App is an all-in-one app, allowing users to search, watch short videos and read headline news, quickly find long-tail information, make direct purchases from Smart Mini Program, play mini games and read online literature, just to name a few. Baidu App also offers voice search, augmented reality search and visual search as well as OCR translation.

Our focus on growing in-app search traffic and expanding our content and services ecosystem is proving to be the right strategy. In addition to growing feed time spend, Baidu App saw search traffic growing in the mid-teens over last year.

Turning to monetization. Our focus to grow in-app traffic build an ecosystem for content, services and vertical offerings, and improve user relationship management will provide tremendous potential to improving recommendation relevancy and native app-like experience for our users as well as continuously improving marketing ROI for our customers.

For example, Jia.com, a home interior-design e-commerce platform, saw leads conversion to in-person consultation from Baidu’s Smart Mini Program increases nearly 30% compared to its HTML5 site. Our focus on in-app traffic and user relationship management will also allow us to provide CRM offering to our customers which will strengthen our relationships with our customers and open a large opportunity for future growth. We are also expanding Baidu’s online — Baidu’s advertising partner network to over 1.5 million out-of-home digital screens, covering 362 cities across 31 provinces.

Turning to DuerOS. Our DuerOS voice assistant continues to lead in China with installed base reaching 275 million, increasing 279% year-over-year and monthly voice queries reaching 2.37 billion, increasing 817% year-over-year in March. We are expanding the service offering in the DuerOS skills store with wide-ranging genre from watching short videos and long videos to playing online games, from listening to children’s stories to setting up tools to help users find their phone. App developers are finding that DuerOS skills, or voice-enabled apps, can translate into great user experience and user retention. For example, almost 40% of live video Douyu skill’s monthly active users on DuerOS continue to use the skill 4 months after initial sign on, which shows incredible user stickiness.

DuerOS first-party device and the breakout quarter, with first quarter Xiaodu series smart devices surpassing the unit sales of all 2018. OEM manufacturers are adopting DuerOS voice assistant as well. For example Skyworth, a leading smart TV manufacturer in China, recently switched the wake word on its TVs to Xiaodu Xiaodu, and Huawei tablets come with tablet mounts that convert the tablet into speech-recognition-enabled DuerOS smart devices.

On the auto front, Chery Automobile EXEED sedans are selling with DuerOS infotainment system preinstalled. DuerOS for Apollo has received a very positive initial customer feedback, particularly for facial recognition capabilities, to activate personalized settings, online payment and AR navigation. On the service front, over 190 Smart Mini Programs from Baidu App and skills from DuerOS home devices are now available in the skills store of DuerOS for Apollo, including iQIYI, E Designated Driving and GeekPark.

Turning to Apollo. The Beijing Transportation Commission recognized Apollo as the leader in autonomous driving, amassing over 10x the test miles of the next industry player, according to the commission’s 2018 Beijing Autonomous Driving Vehicles Road Test Report issued in March. The report measures safety and quality control of autonomous driving and bases its assessment on test miles monitored by the commission as compared to self-reported miles used by other autonomous driving reports. The Apollo developer community has now expanded to 15,000 strong. We are pleased to see that Apollo autopilot technologies have been adopted in many innovative scenarios, including street cleaning, goods delivery and shuttle services.

Transportation related to vehicles and road infrastructure represent approximately 11% of China’s GDP, according to the National Bureau of Statistics. With the Chinese government open to experimenting the smart transportation to improve urban living, we see an opportunity to partner with local governments to implement autonomous driving and smart transportation solution across China. We are working with municipalities to pilot robotaxi by the second half of this year. We are also working with Haidian District of Beijing to develop an AI-powered city brain to improve municipal services. Our entrance into smart transportation is opening doors to the government market and potentially expanding into broader smart city solutions to help Chinese cities modernize.

Turning to Baidu Cloud. Our cloud business continued to exhibit strong revenue momentum, growing triple digits year-over-year. Last October, we discussed implementing Baidu AI call center solution at one of top telecom operators in China. Baidu Brain is now handling approximately 5.6 million calls per month. Our telecom operator customer was pleased with Baidu’s AI enterprise solution and has signed up to expand Baidu AI call center solutions for half of their call center nationwide. In addition, we are receiving orders from enterprise customers in other industries, such as financial services, airline and energy.

At the Baidu Cloud Internet Summit in April, we released a stack of enterprise solutions for companies working with videos. We also released a library of new toolkit that can help developers create, edit, analyze and manage video content using Baidu AI, including content-adaptive encoding, fully interactive augmented reality and a comprehensive video AI model training platform. These solution stacks and toolkit will benefit customers in the video, education, gaming and health care industries.

Baidu Cloud is focused on using Baidu AI to solve our customer problems and increase corporate productivity. We differentiate through heavy investment in AI and our support for open source deep learning platform.

According to the World Intellectual Property Organization 2019 Artificial Intelligence Report, Baidu is the leading company globally in patent application for deep learning. Baidu is also the only Chinese Internet company listed among the top 30 AI patent applicants worldwide.

In March, developer downloads of PaddlePaddle, our leading Chinese open source deep learning platform, was up 40% sequentially. PaddlePaddle helps developers to enhance and streamline their deep learning development modules and drive greater adoption of deep learning in real-world applications on Baidu Cloud.

Turning to iQIYI. IQIYI continues to see strong subscriber growth with membership reaching 96.8 million, increasing 58% year-over-year and further strengthening iQIYI’s foundation to offer blockbuster original entertainment content.

A quick word on Baidu’s corporate social responsibility. Baidu data centers have obtained more than 400 technology and innovation patents in China and abroad. For example, our Yangquan Cloud Computing center, the first double 5A energy-saving data center in China, is widely regarded as an industry role model with an annual PUE of 1.09, reaching to a top-tier level at global standard, much higher than the average level in China. In 2018, through technology innovation and clean energy adoption, Yangquan Cloud Computing center reduced carbon dioxide emission by 177,000 tons. For more information on Baidu’s corporate social responsibility effort, please refer to our newly published CSR report, the website link to which can be found in our earnings release.

With that, let me turn the call over to Herman to go through the financial highlights.

——————————————————————————–

Herman Yu, Baidu, Inc. – CFO [4]

——————————————————————————–

Thanks Robin. Hello, everyone. Welcome to Baidu’s first quarter 2019 call. Before I begin the financial review, let me make a few comments. All monetary amounts used in my discussion are in renminbi, and all growth rates assume year-over-year growth, unless stated otherwise.

For the first quarter 2019, total revenues reached CNY 24.1 billion, up 15%, or 21%, excluding spinoff revenue of CNY 1.1 billion. At the same time, we incurred net loss of CNY 327 million due to our investment in CCTV’s Chinese New Year Eve Gala marketing campaign as well as increased loss from iQIYI. Our net margin also decreased with our focus on in-app growth over TAC revenue growth. In-app growth changes our business model requiring us to spend more heavily upfront on marketing, for app installs and user education, while monetization occurs over the lifetime of the users, TAC revenue, on the other hand, is earned in the quarter TAC is incurred. Investing in AI and leveraging user insight to improve user relationship management, can offer us greater growth potential than Union traffic. Revenue from TAC declined in the first quarter, as bidding for Union traffic, especially on mobile, continues to be competitive. Our goal for Union traffic is to maximize profit, unlike some of our peers who’s bidding on negative margins. The fierce market competition for TAC is also weighing down our Q1 margin. The bright side on TAC is that Baidu’s ad network is expanding into out-of-home digital screens, which complement Baidu’s omni-marketing strategy to reach our users in multiple ways and is becoming a source of future revenue growth.

As Robin mentioned, we are reviewing our business for opportunities to increase operational efficiency. This process will occur over the next few months, and we will look for opportunity to lessen the currently planned increase of approximately CNY 1 billion in cost structure each quarter for the remaining quarters this year. Nevertheless, our priority remains to strengthen our mobile foundation, including growing our search plus feed apps, and to lead in new AI businesses.

Our Board recently approved USD 1 billion for a stock repurchase. The plan for which is effective until July 1st next year. This is in addition to the current USD 1 billion stock buyback plan in place, approximately USD 500 million of which is available for repurchase until the end of next month.

Overall, we remain optimistic about Baidu’s future. In the past 19 years, we have weathered macro conditions, government policy changes and technological changes, each time returning to growth several quarters later. We’re excited about our current strategy to make the fastest growing areas of content searchable while adding news feed to our distribution channel; to grow our ecosystem and content services and vertical offerings, enabling us to perform user relationship management and to take advantage of our AI capabilities to capture market opportunities, with voice assistant, cloud and autonomous driving, which may cause us to sacrifice short-term profit but optimize sustainable, long-term growth.

Let me now turn to Q1 financial review. Revenue from Baidu Core grew to CNY 17.5 billion or USD 2.6 billion, up 16%, excluding spinoff revenues, which is CNY 0.1 billion lower than the midpoint assumption we had used for our Q1 guidance. During Q1, we saw revenue stream coming from education, retail/e-commerce and services sector, which was partially offset by weakness in health care, online gaming, financial services, primarily due to industry-specific government regulations and our self-directed health care initiative.

At the end of March, the majority of our health care customers have been switched — have switched their ad landing pages onto our platform. Though our structured data initiatives will dampen near-term revenue growth Q2 — in Q2, we’ll have a full quarter impact. We believe better service quality and user experience will strengthen Baidu’s foundation for sustainable long-term growth. Our structured data initiative is generating healthy traffic growth, and we plan to roll out this initiative to other service sectors, such as moving companies, home services and express delivery, to improve the user experience in these areas.

Our new AI services are growing fast, particularly our cloud business, which generated CNY 1.3 billion in revenues in the first quarter, up 133%. Revenue from iQIYI reached CNY 7 billion, growing at a robust rate of 43%. Membership revenue continues to be strong with 9.4 million subscribers added in Q1, driven by premium content and hot originals.

Turning to cost of sales. Excluding stock compensation and intangible asset amortization, cost of sales was CNY 14.6 billion, up 49%. Content cost was up 47% to CNY 6.2 billion, mainly due to iQIYI’s increased investment in content, and to a much lesser extent, investment in feed content.

SG&A expenses. Excluding stock compensation, SG&A expenses were CNY 5.5 billion, up 94%, primarily due to the increase in channel and promotional marketing, mainly for Baidu family of apps, including marketing campaigns around the Chinese New Year Eve as well as increasing personnel-related expenses.

R&D expenses, excluding stock compensation, were CNY 3.5 billion, up 25%, primarily due to an increase in personnel-related expenses. Non-GAAP operating income was CNY 0.4 billion. Non-GAAP operating income for Baidu Core was CNY 2.1 billion, and non-GAAP operating margin for Baidu Core was 12%.

Income tax was CNY 294 million compared to CNY 1.1 billion last year. Lower income taxes mainly due to lower pretax income from Baidu Core.

Non-GAAP net income to Baidu was CNY 967 million, and non-GAAP net margin was 4%. Non-GAAP net income attributable to Baidu Core was CNY 1.8 billion, and non-GAAP net margin for Baidu Core was 10%.

Adjusted EBITDA was CNY 1.8 billion, and adjusted EBITDA margin was 7%. Adjusted EBITDA for Baidu Core was CNY 3.4 billion, and adjusted EBITDA margin for Baidu Core reached 19% in Q1.

As of March 31, 2019, cash and short-term investments were CNY 143.6 billion or USD 21.4 billion. Excluding iQiyi, cash and short-term investments for Baidu Core was CNY 125.7 billion or USD 18.7 billion.

Operating cash flow was CNY 1.7 billion. Operating cash flow for Baidu Core was CNY 1.3 billion or USD 191 million. Total headcount of Baidu Core was approximately 32,600, down 1% year-over-year.

Turning to second quarter guidance. We expect total revenues to be between CNY 25.1 billion and CNY 26.6 billion, representing negative 3% to 2% increase year-over-year, or 1% to 6% increase year-over-year, excluding spinoff revenues of CNY 1 billion for Q2. Excluding spinoff revenues, our guidance assumes Baidu Core will grow between negative 2% and 4% increase year-over-year. These forecasts are current and preliminary view and are subject to change.

I will now open the call to questions. Thank you.

——————————————————————————–

Sharon Ng, Baidu, Inc. – Director of Investor Relations Department [5]

——————————————————————————–

Operator, first question, please.

================================================================================

Questions and Answers

——————————————————————————–

Operator [1]

——————————————————————————–

(Operator Instructions) Your first question comes from the line of Alicia Yap of Citigroup.

——————————————————————————–

Alicia Yap, Citigroup Inc, Research Division – MD and Head of Pan-Asia Internet Research [2]

——————————————————————————–

I have a question regarding the ads outlook and also the second quarter guidance. Could you share with us, firstly, the reasons why Mr. Hailong Xiang decided to leave? Is it related to your young manager promotion program? Will these be any potential short-term impact to the search team morale or any impact to sentiment among your key search advertisers? Based on your second quarter guidance, Herman, I just wanted to clarify, you mentioned earlier. So Core Baidu revenues is implying about a negative 2% to plus 4% year-over-year, right? So does this has anything to do with the departure of Mr. Xiang? Or is it mainly the macro softness impact on the apps demands? And how long do you expect this challenging outlook to last?

——————————————————————————–

Robin Li, Baidu, Inc. – Co-Founder, Chairman & CEO [3]

——————————————————————————–

Alicia, let me answer the first part of your question. Ms. Xiang resigned for personal reasons, and Shen Dou is promoted to Senior Vice President who’s in charge of our mobile ecosystem which was formerly named search company. I think Dou is very strong. He has proven track record in managing a lot of businesses, most recently our feed business has been growing very fast. And if you look at the user front, we have already built a relatively strong mobile ecosystem with a large apps of super apps with very high DAU, the time spent is growing, search traffic in the app is growing, and our organic content and services from Baijiahao as well as Smart Mini Programs are all growing. So going forward, we are confident that, that when we build the stronger sales team and stronger monetization capability, we should be right on track for higher growth.

——————————————————————————–

Herman Yu, Baidu, Inc. – CFO [4]

——————————————————————————–

Alicia, this is Herman. Yes, I wanted to just confirm what you noticed. With regards to our guidance for Q2, we are assuming that Baidu Core will be growing approximately negative 2% to positive 4% year-over-year for Q2. And you asked us what were the considerations that we have for Q2, I think several aspects. As Robin mentioned in his prepared remarks, let me kind of summarize the key factors. I think one is macro factors that we talked about. Number two is we started seeing, after Chinese New Year, that there’s been an increase in inventory, ad inventory release in the market. And as a result, that has been impacting our CPM. So obviously, we’re going to be focusing on improving our monetization capabilities going forward. Thirdly, we talked about our health care initiative. I recall in November last year, we said that we’re going to be doing this health care initiative for structured data. We are still on track. We’re moving everyone, health care trying to move everyone onto Baidu’s landing page. So that was done in March. So you’re going to absorb a full quarter of that impact in Q2. We think over the long run, since we now understand the users better, we will understand their behavior as we’re navigating on the structure site. We think that, albeit in the long term, we can continue to improve the experience we’re seeing in traffic from that initiative actually increasing growth. So I think over time, we’re going to be able to improve our conversion. And lastly, what we’re assuming for the policies that have impacted us, whether it’s on gaming, whether it’s on financial services, on real estate, on auto and so forth, will continue to be this way. So if these industry-specific policies improve, that will be upside for us, okay?

——————————————————————————–

Operator [5]

——————————————————————————–

Your next question comes from the line of Eddie Leung of BofA Merrill Lynch.

——————————————————————————–

Eddie Leung, BofA Merrill Lynch, Research Division – MD in Equity Research and Analyst [6]

——————————————————————————–

Just wondering if you could share more color with us on the growth rate you have seen some of your key verticals, client or advertiser verticals regarding the softness in the first quarter and second quarter. Specifically, are we seeing an across-the-board weakness with a similar magnitude? Or are we seeing a couple of verticals dragging down the overall growth rate? And if so, could you talk a little bit about that field verticals? I suppose maybe vertical in financial services. Anything along the line would be very helpful.

——————————————————————————–

Herman Yu, Baidu, Inc. – CFO [7]

——————————————————————————–

Eddie, this is Herman. Let me answer that. As we — as I mentioned in the prepared remarks, several industries probably have the slowest growth rate, health care, online gaming and financial services. I think these 3 will probably be the hardest hit on a year-over-year basis. Beyond that, as we mentioned, this is a macro impact. It’s impacting our CPMs. So as a result, you’re seeing a slowdown growth rate in many industries beyond those 3 that I mentioned.

——————————————————————————–

Operator [8]

——————————————————————————–

Your next question comes from the line of Gregory Zhao from Barclays.

——————————————————————————–

Gregory Zhao, Barclays Bank PLC, Research Division – VP [9]

——————————————————————————–

One follow-up of your guidance please. We’re looking to the business segment of Baidu Core, we know they are search, feed and cloud. But I just want to understand the group’s trend of each individual segment and what’s the driver for each segment in your Q2 guidance? And also a quicker follow-up. During your prepared remarks, you mentioned that the efficiency improvement plan and you also mentioned the RMB 1 billion. So is that for — is that on saving quarter-by-quarter? Or more investment quarter-by-quarter?

——————————————————————————–

Herman Yu, Baidu, Inc. – CFO [10]

——————————————————————————–

Greg, this is Herman. So 2 questions, one on drivers of search, feed and cloud. So I’ll go first and Robin can add. So I think as we mentioned earlier, for our Q2 guidance, the 4 factors that we’re looking at right now that’s impacting our growth rate, the macro, our initiative for structured data and the fact that the market is seeing significant increase in supply of inventory, and some industry-specific policies, I think those would give you a good indication of the drivers. Obviously, we’re going to be focusing monetization. We talked about how we’re going to focus on CRM. We’re going to focus on other initiatives to increase our CPMs so that we can become more competitive. I think those are the drivers for advertising. And then we break out specifically for cloud that it’s CNY 1.3 billion, growing at 133% year-over-year. I think the driver for us is to continue to expand into different industries and to replicate a lot of the solutions, enterprise solutions that we have so that it’s much faster to grow with existing solutions rather than having to customize new solutions. So that’s kind of our strategy to accelerate the growth of our cloud business. And then with regards to cost structures, we originally had planned at the beginning of the year to increase our cost of sales and operating expenses by CNY 1 billion per quarter. So the increase incrementally CNY 1 billion quarter-over-quarter for the remaining quarters of this year. We have started looking at each of these areas. Our goal is to decrease the increase sequentially through our business reviews.

——————————————————————————–

Robin Li, Baidu, Inc. – Co-Founder, Chairman & CEO [11]

——————————————————————————–

Yes, let me give you a little more color on the growth drivers for search, feed and cloud. Cloud, like I mentioned, was growing by a triple digit, a fast growth rate. For search and feed, like I mentioned before, the DAU, the time spent and page views are all growing in the Baidu app. So we have a relatively solid foundation on the user front. But on the customer front, we need to get more sophisticated, especially when we have all the user information in our native app. We can actually shifting our focus from managing traffic to managing users and customers. For example, when a customer switched their landing page from HTML5 one to Smart Mini Program, we typically see that the conversion improved by 30% to 50%. And the traditional Baidu online marketing business is — you already call it pay for performance. And when the performance improved, let’s say, the 30% to 50%, we certainly should be able to take a cut from that. And that should just — the initial result, we believe that there is still a lot of room to improve. And the reason for that is that in the same app, when we have the user information, their mailing address, their demographic information and their telephone number, they don’t need to fill out a lot of new forms before they place an order from the Baidu App. This will significantly improve the conversion rate and provide a more native app-like experience for both the users and online customers. And this kind of trend, I think we’ll be able to last for quite a few years and will drive our growth for the core business.

——————————————————————————–

Operator [12]

——————————————————————————–

Your next question comes from the line of Juan Lin of 86Research.

——————————————————————————–

Juan Lin, 86Research Limited – Research Analyst [13]

——————————————————————————–

So my question is also on core business. As you mentioned, the DAU of Baidu App was up by 28% year-over-year. And now that we’re already in the middle of the second quarter, I wonder if you have seen a similar DAU improvement on a year-over-year basis for the second quarter? Also, there’s the disparity between user growth and the revenue growth for the core business obviously. So I wonder when should we see user growth start to contribute to the core revenue growth? Second question is out-of-home screen and OTV. I wonder how big is the online TV and offline screen advertising business? What is the margin profile of these 2 business lines? And what is the growth trend for these 2 business lines?

——————————————————————————–

Robin Li, Baidu, Inc. – Co-Founder, Chairman & CEO [14]

——————————————————————————–

It’s going to be a gradual process to see user growth turning into revenue growth. Like I mentioned, there are a lot of work we can do and we already started to do this kind of work. We will focus more on managing our users and managing our customers so that the conversion will continue to improve and so that customers are willing to pay more going forward. And on the outdoor digital screens, those are not really offline screens, those are online screens. We can leverage that as well as our mobile, in-home devices, auto, in auto services to do omni-marketing. And when we are able to connect that the users and customers across all different platforms, I believe that, that customers will be better served.

——————————————————————————–

Juan Lin, 86Research Limited – Research Analyst [15]

——————————————————————————–

So just a quick follow up on the user growth for the second quarter. So far, have you seen a similar year-over-year growth in terms of core mobile Baidu App in terms of user growth so far in the second quarter?

——————————————————————————–

Robin Li, Baidu, Inc. – Co-Founder, Chairman & CEO [16]

——————————————————————————–

I would say that our user growth continued to be healthy, but we typically do not give up on a daily basis update on the DAU growth.

——————————————————————————–

Herman Yu, Baidu, Inc. – CFO [17]

——————————————————————————–

Yes, just to address your plan onto that, Juan. The 28% year-over-year is what we reported for March. So if you’re implying that was this impacted by the Chinese New Year. Chinese New Year, as you know, happened in the first half of February. So whatever that impact, it does have a branding effect that allows our users to continue to come back even after Chinese New Year. So I think March is a good indication of the fact that the marketing campaign worked and that people remember and that people are willing to come back to Baidu App.

——————————————————————————–

Operator [18]

——————————————————————————–

Your next question comes from the line of Piyush Mubayi of Goldman Sachs.

——————————————————————————–

Piyush Mubayi, Goldman Sachs Group Inc., Research Division – MD [19]

——————————————————————————–

I wonder whether you could split your guidance and the slowdown that you’re seeing into 2019 second quarter between macro, the release of new inventory that you talked about, Herman, and the issues that you’re facing with the health care initiated for structured data. If you could split it? And then if you could talk through if at all possible into when we could see the recovery for the force, in particular the health care factor into the second half of 2019. And my second question is concerning your gross margins which, for the core for the first time appear to be in the 50s, it looks like based on our numbers, about 35%, a number we haven’t seen in a long time. Should we expect that to be the run rate into the second quarter and the rest of the year?

——————————————————————————–

Herman Yu, Baidu, Inc. – CFO [20]

——————————————————————————–

Piyush, can you repeat that again? What you’re implying for 50-some percent?

——————————————————————————–

Piyush Mubayi, Goldman Sachs Group Inc., Research Division – MD [21]

——————————————————————————–

The growth and profit margin — the gross profit margin for the second — for the first quarter looks like it’s 55%, which is lower than the 64%, 68% and 70% we’ve seen in 4Q, 3Q and second quarter for the core business. So that’s a run rate we should be factoring in as we think through the rest of 2019.

——————————————————————————–

Herman Yu, Baidu, Inc. – CFO [22]

——————————————————————————–

Yes. Good question. I think for our gross margin, the 2 factors you want to consider, number one, is the fact that we’re selling smart devices into homes DuerOS so hardware sales, as you know, margin is very thin. So as a result, the more hardware sales we’re selling. And as we mentioned on the prepared remarks that Q1 sales was basically equivalent of all of last year. So while on the one hand, the sale is doing really well, it’s going to impact on margin. And secondly, as I also mentioned in the prepared remarks of our TAC revenues, so you’re seeing TAC revenues on the whole declining. At the same time, you’re seeing the cost of TAC revenue growing closer to 40% year-over-year in Q1. So these are 2 factors. I would expect going into Q2, with similar trend where you have Q1 TAC revenue was declining. Going into Q2, I don’t think the TAC revenue is going to grow too strongly at all. On the other hand, I expect smart devices to continue to do well.

——————————————————————————–

Robin Li, Baidu, Inc. – Co-Founder, Chairman & CEO [23]

——————————————————————————–

And yes, on the headwinds you mentioned macro, new inventory and structure of the data. It’s very hard for us to accurately split the percentage amount of these factors, but I think that the general direction is that for the structured data, we have built a good base to improve from. There are a lot of levers we can pull to improve the conversion, improve the user experience. When those content are structured and hosted on Baidu, we’re essentially using the Baidu name to endorse those institutions. So longer term, the conversion should get better and better. For the new inventory, new ad inventory on the market, we saw a flash of that over the past — earlier this year, but my expectation is that it’s not going to get worse, but it will probably stay the same for the rest of the year. And for macro, there are still a lot of uncertainties that we really do not have accurate read.

——————————————————————————–

Herman Yu, Baidu, Inc. – CFO [24]

——————————————————————————–

Yes. And I wanted to add a few more points on structured data initiatives. I think this is an area where I think it’s good for Baidu. I think this is something that, in the short term, is going to impact on margin, but long term, this builds a solid foundation for us to grow. For several reasons because what you’re doing is you’re taking over the landing page, right? So you put on Baidu technologies. So in the past, business model is someone searches, we send them to an H5 site and we no longer know what the user does. And then there could be a lot of false advertising on the landing page. We’re now requiring the contents beyond Baidu, and we’re using our technology to screen it. So we gained several things. Number one is we get to understand the users more, and at the same time, we request, for example, for these medical institutions that they have the physicians’ bio which will allow us in the future to do verification. We try to do rating because we now have more relationship with this particular vendor, and we would ask for things like certain business license and so forth. We’re able to — because it’s on our technical platform, we can do screening using our AI. We’re adding social elements to this so that people can comment on and these rate these vendors and so forth. So you can think of how structured it is very similar, for example, the e-commerce where once you have the rating system, you have social and then being able to monitor what’s on their site, improve the user experience, you can see that also, for example, with hotels on travel sites and so forth. So I think over time, you’re going to actually see us build a solid foundation for revenue improvement because we think there’s going to be better conversion. Already, we’re seeing traffic in the medical health care areas improve as a result of doing this.

——————————————————————————–

Operator [25]

——————————————————————————–

Your next question comes from the line of Grace Chen from Morgan Stanley.

——————————————————————————–

Grace Chen, Morgan Stanley, Research Division – Equity Analyst [26]

——————————————————————————–

I have a question about the Ad business. Several Internet companies have mentioned about the challenges in online advertisement business in the first quarter. And I think they all point out to similar reasons, including macro regulations and supply increase. So I’m wondering, apart from these 3 common factors, are there any other specific — company-specific factors that would differentiate Baidu to perform better or face more challenges in terms of the online ad revenue in the following quarters versus your peers.

——————————————————————————–

Robin Li, Baidu, Inc. – Co-Founder, Chairman & CEO [27]

——————————————————————————–

I think when we look at the changes in the marketplace, I think that they’re out of control like macro, but they are — so we pretty much focus more on things that we have control of, which is how do we improve the user experience and customer experience, and therefore, improve conversion. Like I mentioned before, by replacing HTML5 sites with Smart Mini Program, we can typically improve the conversion by 30% to 50%. And by hosting the landing page using structured data, we’re basically endorsing the customers with the Baidu brand, and we’re also providing third party more objective comments about those advertising materials as well in the long run improve the conversion. And when conversion is improved, the revenue should follow. I think this is more Baidu specific and we have better control of.

——————————————————————————–

Herman Yu, Baidu, Inc. – CFO [28]

——————————————————————————–

Yes, Grace, and I think the other question you were asking or you’re implying, from my understanding, is you’re comparing us to our peers and you’re asking us probably on the growth rate. I think in addition to those 3 areas, number one is our — we span across many industries. Secondly, I think is the structured data we talked about various ads on our platform that we want to — we’ve been doing this starting late last year that we want to actually ensure that we have a better user experience. So as a result, it’s a matter of derisking the properties on Baidu so that in the future we have a more solid foundation to grow. So I think structured data initiative fits into that, and that’s probably something that we’re doing that a lot of peers are not doing. And as a matter of fact, by doing this, some of our peers probably would benefit from some of the advertisers that we selectively reject. But what that means is that you have short-term growth and then you also have a higher risk in the future. So we think that, again, our focus right now is how do we maximize our long-term growth, how do we improve the healthiness of our advertising and so forth.

——————————————————————————–

Operator [29]

——————————————————————————–

Your next question comes from the line of Jerry Liu of UBS.

——————————————————————————–

Jerry Liu, UBS Investment Bank, Research Division – Co Head of HK and China Internet Research [30]

——————————————————————————–

First, just following on with the advertising discussion, I want to ask about 2 of the verticals we discussed. One’s health care and one’s online gaming. Could we see some improvement in ad demand here potentially health care as we move all the landing pages and in online gaming as the regulatory approvals are restarting? And then secondarily, just a clarification on costs. Do we — so net of the investments and the cost savings, how do we see OpEx trending for the rest of the year?

——————————————————————————–

Herman Yu, Baidu, Inc. – CFO [31]

——————————————————————————–

Hi Jerry, so a couple questions in terms of are there upsides to the industries specifically for health care and gaming. I think we talked about the health care. I think the key driver there is we switched the majority of our medical advertisers at the end of Q1. So in Q2, we’re going to have a full quarter hit, because typically when you switch those sites, the users come to our site, they have to get used to it, and it will take a while for them to get used to it to get the conversion to where we are before the switch. And then I think over time, because traffic in this area searches are increasing, it shows that user experience are getting better, and we are now getting better understanding of what the users do after they go to the landing page. So I think over time, using our technology and our experience, we should be able to improve conversion. So I think once we can do that, you’ll be able to see that growth will come back in health care. With regards to gaming, I think you’re right. I think for all the industry-specific policies, it’s going to be in a similar fashion that once, for example, gaming license you’re seeing more and more getting approved that’s going to cause the game developers to be able to develop new games. So whenever they develop new games, they want to launch new games and so forth. I think that revenue is going to come back. The other factors that we talked about in our prepared remarks is the fact that there’s government economic stimulus right now, including a CNY 2 trillion for tax breaks and for other benefits of related costs. So those type of things, we didn’t factor in how that’s going to impact our business because it’s hard to quantify, but if those things start to take toll, that is a potential upside. With regards to our cost structure, as I mentioned, we currently plan about increasing CNY 1 billion per quarter for the remainder of the year. What we’re doing is going to our businesses to see how much we can reduce that. So I think ultimately, you’re probably going to still see a sequential increase, but we would like to not increase at the rate of CNY 1 billion per quarter. Okay, I hope that helps.

——————————————————————————–

Operator [32]

——————————————————————————–

Your last question comes from the line of Wendy Huang from Macquarie.

——————————————————————————–

Wendy Huang, Macquarie Research – Head of Asian Internet and Media [33]

——————————————————————————–

First is a follow-up on the structured data for the health care. So Herman, you mentioned earlier the full quarter impact will be seen in second quarter. So does that mean from the Q3 onwards, we should actually see the growth, actually, both sequentially as well as year-over-year? And also, has this change in the structured data for the health care actually affected the number of advertisers in the health care industry? And also regarding the advertiser side. So are we seeing the lower spending from advertiser each? Or are we seeing actually the less amount of the smaller advertisers because that actually is dying down due to a macro?

——————————————————————————–

Herman Yu, Baidu, Inc. – CFO [34]

——————————————————————————–

Hi Wendy, that’s a mouthful of question there. With regards to health care, it’s hard for me to predict going into our Q3. So my initial reaction would be that if Q2 is going to have a full quarter impact, we’ll probably not going to be able to learn enough about the users and change the user behaviors and so forth in just a matter of a couple of months. It probably will take us a little bit longer. I can’t say at this point when that turning point is going to be, but I don’t think it’s going to be a few years, okay? On the other hand, with regards to a number of customers, yes, as we’re being more selective on the quality of our customers, that’s going to cause us to actually decrease the number of customers for our product health care but I think it’s a good thing. As I mentioned, it’s a matter of do you want to derisk your properties and have a good foundation of the future growth? We’ve always believed maximizing our future growth is where we want to be. And you’re going to some of our peers picking up these probably unhealthy advertisers and probably showing near-term growth. But at the same time, that comes with the risk and regulations and so forth. With regards to are we seeing ARPU with our industries and so forth, I think in general, as I mentioned, this is a macro impact. So we’re seeing ARPU, it depends on which industries, which areas. So it’s not, in the past, on a year-over-year basis, we’ll probably see ARPU increase across the board. I think this is a quarter, especially when we go into Q2, where it’s going to be mixed. It’s hard to have 1 generalization for all the industries. I think right now, with CPM not growing as fast as we had hoped for after Chinese New Year, then we’re going to have mixed results across the board. So I think fundamentally, for us what we can control is improve our monetization capabilities to focus on how do we become better versus our peers. And that’s what we’ve been starting to do.

——————————————————————————–

Operator [35]

——————————————————————————–

We are now approaching the end of the conference call. Thank you for your participation in today’s conference. You may now disconnect. Good day.

Canada lacks laws to tackle problems posed by artificial intelligence: Experts – Global News

The role of artificial intelligence in Netflix’s movie suggestions and Alexa’s voice commands is commonly understood, but less known is the shadowy role AI now plays in law enforcement, immigration assessment, military programs and other areas.

Story continues below

Despite its status as a machine-learning innovation hub, Canada has yet to develop a regulatory regime to deal with issues of discrimination and accountability to which AI systems are prone, prompting calls for regulation — including from business leaders.

“We need the government, we need the regulation in Canada,” said Mahdi Amri, who heads AI services at Deloitte Canada.

READ MORE: How Canadian AI is kick-starting new discoveries in health and science

The absence of an AI-specific legal framework undermines trust in the technology and, potentially, accountability among its providers, according to a report he co-authored.

“Basically there’s this idea that the machines will make all the decisions and the humans will have nothing to say, and we’ll be ruled by some obscure black box somewhere,” Amri said.

Robot overlords remain firmly in the realm of science fiction, but AI is increasingly involved in decisions that have serious consequences for individuals.

Since 2015, police departments in Vancouver, Edmonton, Saskatoon and London, Ont. have implemented or piloted predictive policing _ automated decision-making based on data that predicts where a crime will occur or who will commit it.

WATCH: A.I. fuels unprecedented growth in health and science

The federal immigration and refugee system relies on algorithmically-driven decisions to help determine factors such as whether a marriage is genuine or someone should be designated as a “risk”, according to a Citizen Lab study, which found the practice threatens to violate human rights law.

AI testing and deployment in Canada’s military prompted Canadian AI pioneers Geoffrey Hinton and Yoshua Bengio to warn about the dangers of robotic weapons and outsourcing lethal decisions to machines, and to call for an international agreement on their deployment.

“When you’re using any type of black box system, you don’t even know the standards that are embedded in the system or the types of data that may be used by the system that could be at risk of perpetuating bias,” said Rashida Richardson, director of policy research at New York University’s AI Now Institute.

READ MORE: Google beefs up privacy measures — but some experts aren’t impressed

She pointed to “horror cases,” including a predictive policing strategy in Chicago where the majority of people on a list of potential perpetrators were black men who had no arrests or shooting incidents to their name, “the same demographic that was targeted by over-policing and discriminatory police practices.”

Richardson says it’s time to move from lofty guidelines to legal reform. A recent AI Now Institute report states federal governments should “oversee, audit, and monitor” the use of AI in fields like criminal justice, health care and education, as “internal governance structures at most technology companies are failing to ensure accountability for AI systems.”

Oversight should be divided up among agencies or groups of experts instead of hoisting it all onto a single AI regulatory body, given the unique challenges and regulations specific to each industry, the report says.

In health care, AI is poised to upend the way doctors practice medicine as machine-learning systems can now analyze vast sets of anonymized patient data and images to identify health problems ranging from osteoporosis to lesions and signs of blindness.

Carolina Bessega, co-founder and chief scientific officer of Montreal-based Stradigi AI, says the regulatory void discourages businesses from using AI, holding back innovation and efficiency _ particularly in hospitals and clinics, where the implications can be life or death.

WATCH: How the future will be impacted by artificial intelligence

“Right now it’s like a grey area, and everybody’s afraid making the decision of, ‘Okay, let’s use artificial intelligence to improve diagnosis, or let’s use artificial intelligence to help recommend a treatment for a patient,’” Bessega said.

She is calling for “very strong” regulations around treatment and diagnosis and for a professional to bear responsibility for any final decisions, not a software program.

Critics say Canada lags behind the U.S. and the EU on exploring AI regulation. None has implemented a comprehensive legal framework, but Congress and the EU Commission have produced extensive reports on the issue.

“Critically, there is no legal framework in Canada to guide the use of these technologies or their intersection with foundational rights related to due process, administrative fairness, human rights, and justice system transparency,” states a March briefing by Citizen Lab, the Law Commission of Ontario and other bodies.

READ MORE: Canada’s use of artificial intelligence in immigration could lead to break of human rights: study

Divergent international standards, trade secrecy and algorithms’ constant “fluidity” pose obstacles to smooth regulation, says Miriam Buiten, junior professor of law and economics at the University of Mannheim.

Canada was among the first states to develop an official AI research plan, unveiling a $125-million strategy in 2017. But its focus was largely scientific and commercial.

In December, Prime Minister Trudeau and French President Emmanuel Macron announced a joint task force to guide AI policy development with an eye to human rights.

Minister of Innovation, Science and Economic Development Navdeep Bains told The Canadian Press in April a report was forthcoming “in the coming months.” Asked whether the government is open to legislation around AI transparency and accountability, he said: “I think we need to take a step back to determine what are the core guiding principles.

“We’ll be coming forward with those principles to establish our ability to move forward with regards to programming, with regards to legislative changes — and it’s not only going to be simply my department, it’s a whole government approach.”

The Treasury Board of Canada has already laid out a 119-word set of principles on responsible AI use that stress transparency and proper training. The Department of Innovation, Science and Economic Development highlighted the Personal Information Protection and Electronic Documents Act, privacy legislation that applies broadly to commercial activities and allows a privacy commissioner to probe complaints.

“While AI may present some novel elements, it and other disruptive technologies are subject to existing laws and regulations that cover competition, intellectual property, privacy and security,” a department spokesperson said in an email.

As of April 1, 2020, government departments seeking to deploy an automated decision system must first conduct an “algorithmic impact assessment” and post the results online.

Canada lacks laws to tackle major problems posed by artificial intelligence, experts warn – The Globe and Mail

The role of artificial intelligence in Netflix’s movie suggestions and Alexa’s voice commands is commonly understood, but less known is the shadowy role AI now plays in law enforcement, immigration assessment, military programs and other areas.

Despite its status as a machine-learning innovation hub, Canada has yet to develop a regulatory regime to deal with issues of discrimination and accountability to which AI systems are prone, prompting calls for regulation – including from business leaders.

“We need the government, we need the regulation in Canada,” said Mahdi Amri, who heads AI services at Deloitte Canada.

Story continues below advertisement

The absence of an AI-specific legal framework undermines trust in the technology and, potentially, accountability among its providers, according to a report he co-authored.

“Basically there’s this idea that the machines will make all the decisions and the humans will have nothing to say, and we’ll be ruled by some obscure black box somewhere,” Mr. Amri said.

Robot overlords remain firmly in the realm of science fiction, but AI is increasingly involved in decisions that have serious consequences for individuals.

Since 2015, police departments in Vancouver, Edmonton, Saskatoon and London, Ont. have implemented or piloted predictive policing – automated decision-making based on data that predicts where a crime will occur or who will commit it.

The federal immigration and refugee system relies on algorithmically driven decisions to help determine factors such as whether a marriage is genuine or someone should be designated as a “risk,” according to a Citizen Lab study, which found the practice threatens to violate human rights law.

AI testing and deployment in Canada’s military prompted Canadian AI pioneers Geoffrey Hinton and Yoshua Bengio to warn about the dangers of robotic weapons and outsourcing lethal decisions to machines, and to call for an international agreement on their deployment.

“When you’re using any type of black box system, you don’t even know the standards that are embedded in the system or the types of data that may be used by the system that could be at risk of perpetuating bias,” said Rashida Richardson, director of policy research at New York University’s AI Now Institute.

Story continues below advertisement

She pointed to “horror cases,” including a predictive policing strategy in Chicago where the majority of people on a list of potential perpetrators were black men who had no arrests or shooting incidents to their name, “the same demographic that was targeted by over-policing and discriminatory police practices.”

Ms. Richardson says it’s time to move from lofty guidelines to legal reform. A recent AI Now Institute report states federal governments should “oversee, audit, and monitor” the use of AI in fields like criminal justice, health care and education, as “internal governance structures at most technology companies are failing to ensure accountability for AI systems.”

Oversight should be divided among agencies or groups of experts instead of foisting it all onto a single AI regulatory body, given the unique challenges and regulations specific to each industry, the report says.

In health care, AI is poised to upend the way doctors practise medicine as machine-learning systems can now analyze vast sets of anonymized patient data and images to identify health problems ranging from osteoporosis to lesions and signs of blindness.

Carolina Bessega, co-founder and chief scientific officer of Montreal-based Stradigi AI, says the regulatory void discourages businesses from using AI, holding back innovation and efficiency — particularly in hospitals and clinics, where the implications can be life or death.

“Right now it’s like a grey area, and everybody’s afraid making the decision of, ‘Okay, let’s use artificial intelligence to improve diagnosis, or let’s use artificial intelligence to help recommend a treatment for a patient,’” Dr. Bessega said.

Story continues below advertisement

She is calling for “very strong” regulations around treatment and diagnosis and for a professional to bear responsibility for any final decisions, not a software program.

Critics say Canada lags behind the U.S. and the EU on exploring AI regulation. None has implemented a comprehensive legal framework, but Congress and the EU Commission have produced extensive reports on the issue.

“Critically, there is no legal framework in Canada to guide the use of these technologies or their intersection with foundational rights related to due process, administrative fairness, human rights, and justice system transparency,” states a March briefing by Citizen Lab, the Law Commission of Ontario and other bodies.

Divergent international standards, trade secrecy and algorithms’ constant “fluidity” pose obstacles to smooth regulation, says Miriam Buiten, junior professor of law and economics at the University of Mannheim.

Canada was among the first states to develop an official AI research plan, unveiling a $125-million strategy in 2017. But its focus was largely scientific and commercial.

In December, Prime Minister Justin Trudeau and French President Emmanuel Macron announced a joint task force to guide AI policy development with an eye to human rights.

Story continues below advertisement

Minister of Innovation, Science and Economic Development Navdeep Bains told the Canadian Press in April a report was expected “in the coming months.” Asked whether the government is open to legislation around AI transparency and accountability, he said: “I think we need to take a step back to determine what are the core guiding principles.

“We’ll be coming forward with those principles to establish our ability to move forward with regards to programming, with regards to legislative changes — and it’s not only going to be simply my department, it’s a whole government approach.”

The Treasury Board of Canada has already laid out a 119-word set of principles on responsible AI use that stress transparency and proper training. The Department of Innovation, Science and Economic Development highlighted the Personal Information Protection and Electronic Documents Act, privacy legislation that applies broadly to commercial activities and allows a privacy commissioner to probe complaints.

“While AI may present some novel elements, it and other disruptive technologies are subject to existing laws and regulations that cover competition, intellectual property, privacy and security,” a department spokesperson said in an email.

As of next April 1, government departments seeking to deploy an automated decision system must first conduct an “algorithmic impact assessment” and post the results online.

Experts say Canada lacks laws to tackle life-and-death problems posed by AI – Financial Post

TORONTO — The role of artificial intelligence in Netflix’s movie suggestions and Alexa’s voice commands is commonly understood, but less known is the shadowy role AI now plays in law enforcement, immigration assessment, military programs and other areas.

Despite its status as a machine-learning innovation hub, Canada has yet to develop a regulatory regime to deal with issues of discrimination and accountability to which AI systems are prone, prompting calls for regulation — including from business leaders.

“We need the government, we need the regulation in Canada,” said Mahdi Amri, who heads AI services at Deloitte Canada.

The absence of an AI-specific legal framework undermines trust in the technology and, potentially, accountability among its providers, according to a report he co-authored.

“Basically there’s this idea that the machines will make all the decisions and the humans will have nothing to say, and we’ll be ruled by some obscure black box somewhere,” Amri said.

Robot overlords remain firmly in the realm of science fiction, but AI is increasingly involved in decisions that have serious consequences for individuals.

Since 2015, police departments in Vancouver, Edmonton, Saskatoon and London, Ont. have implemented or piloted predictive policing — automated decision-making based on data that predicts where a crime will occur or who will commit it.

The federal immigration and refugee system relies on algorithmically-driven decisions to help determine factors such as whether a marriage is genuine or someone should be designated as a “risk”, according to a Citizen Lab study, which found the practice threatens to violate human rights law.

AI testing and deployment in Canada’s military prompted Canadian AI pioneers Geoffrey Hinton and Yoshua Bengio to warn about the dangers of robotic weapons and outsourcing lethal decisions to machines, and to call for an international agreement on their deployment.

“When you’re using any type of black box system, you don’t even know the standards that are embedded in the system or the types of data that may be used by the system that could be at risk of perpetuating bias,” said Rashida Richardson, director of policy research at New York University’s AI Now Institute.

She pointed to “horror cases,” including a predictive policing strategy in Chicago where the majority of people on a list of potential perpetrators were black men who had no arrests or shooting incidents to their name, “the same demographic that was targeted by over-policing and discriminatory police practices.”

Richardson says it’s time to move from lofty guidelines to legal reform. A recent AI Now Institute report states federal governments should “oversee, audit, and monitor” the use of AI in fields like criminal justice, health care and education, as “internal governance structures at most technology companies are failing to ensure accountability for AI systems.”

Oversight should be divided up among agencies or groups of experts instead of hoisting it all onto a single AI regulatory body, given the unique challenges and regulations specific to each industry, the report says.

In health care, AI is poised to upend the way doctors practice medicine as machine-learning systems can now analyze vast sets of anonymized patient data and images to identify health problems ranging from osteoporosis to lesions and signs of blindness.

Carolina Bessega, co-founder and chief scientific officer of Montreal-based Stradigi AI, says the regulatory void discourages businesses from using AI, holding back innovation and efficiency — particularly in hospitals and clinics, where the implications can be life or death.

“Right now it’s like a grey area, and everybody’s afraid making the decision of, ‘Okay, let’s use artificial intelligence to improve diagnosis, or let’s use artificial intelligence to help recommend a treatment for a patient,”‘ Bessega said.

She is calling for “very strong” regulations around treatment and diagnosis and for a professional to bear responsibility for any final decisions, not a software program.

Critics say Canada lags behind the U.S. and the EU on exploring AI regulation. None has implemented a comprehensive legal framework, but Congress and the EU Commission have produced extensive reports on the issue.

“Critically, there is no legal framework in Canada to guide the use of these technologies or their intersection with foundational rights related to due process, administrative fairness, human rights, and justice system transparency,” states a March briefing by Citizen Lab, the Law Commission of Ontario and other bodies.

Divergent international standards, trade secrecy and algorithms’ constant “fluidity” pose obstacles to smooth regulation, says Miriam Buiten, junior professor of law and economics at the University of Mannheim.

Canada was among the first states to develop an official AI research plan, unveiling a $125-million strategy in 2017. But its focus was largely scientific and commercial.

In December, Prime Minister Trudeau and French President Emmanuel Macron announced a joint task force to guide AI policy development with an eye to human rights.

Minister of Innovation, Science and Economic Development Navdeep Bains told The Canadian Press in April a report was forthcoming “in the coming months.” Asked whether the government is open to legislation around AI transparency and accountability, he said: “I think we need to take a step back to determine what are the core guiding principles.

“We’ll be coming forward with those principles to establish our ability to move forward with regards to programming, with regards to legislative changes — and it’s not only going to be simply my department, it’s a whole government approach.”

The Treasury Board of Canada has already laid out a 119-word set of principles on responsible AI use that stress transparency and proper training. The Department of Innovation, Science and Economic Development highlighted the Personal Information Protection and Electronic Documents Act, privacy legislation that applies broadly to commercial activities and allows a privacy commissioner to probe complaints.

“While AI may present some novel elements, it and other disruptive technologies are subject to existing laws and regulations that cover competition, intellectual property, privacy and security,” a department spokesperson said in an email.

As of April 1, 2020, government departments seeking to deploy an automated decision system must first conduct an “algorithmic impact assessment” and post the results online.

Witness: Inmate prays, sings during execution – Yahoo News

A man convicted of killing his wife decades ago prayed and sang hymns as he was put to death Thursday in Nashville, Tennessee. Don Johnson, 68, was executed via lethal injection for the 1984 suffocation of his wife, Connie Johnson. (May 17)