Curated deep tech and AI content that humans at Radical are reading and thinking about. Sign up here to have Radical Reads delivered directly to your inbox every week.
1) AI Talent: With a big vision, Toronto founders can attract AI talent from Silicon Valley (Vector Institute and Radical Ventures)
“With nearly 15,000 tech companies, the Toronto-Waterloo corridor is the second largest technology cluster on the continent, boasting a high concentration of AI startups. Venture capital interest in Canadian AI companies is growing, with $658 million invested in 2019, up from $289 million two years prior, the majority of which was invested in Toronto and Montreal-based companies. The last three years have also seen a flood of investment from large organizations, with over 45 corporate AI labs opening or expanding in the region. Ambition, expertise, and energy permeate the corridor. For engineers interested in moving to a city with a flourishing community of high-performing peers and interesting projects, Toronto is a prime destination.
…Meg Lizza, Director of Talent at Radical Ventures, an early-stage venture fund focused on AI and deep tech investments, says, “There’s impact where you’re going into a team and you as an individual contributor will be building something from scratch and impacting the company as a whole,” or doing “something that’s improving the current state of the world.” This means that founders need to sell a big vision, but also plan and speak authentically about the magnitude of impact that a candidate has the opportunity to make if they join the team.”
Radical Commentary: This article is a collaboration between Radical Ventures and the Vector Institute for AI. It describes 5 specific and practical actions to help recruit AI talent.
2) Regulating AI: New Zealand claims world first in setting standards for government use of algorithms (The Guardian)
“The New Zealand charter… was due to launch on Tuesday with 19 government agencies as initial signatories. In it, departments pledge to be publicly transparent about how decision-making is driven by algorithms, including giving “plain English” explanations; to make available information about the processes used and how data is stored unless forbidden by law (such as for reasons of national security); and to identify and manage biases informing algorithms.
Agencies must also consider te ao Māori, or Indigenous, worldviews on data collection — in New Zealand, Māori are disproportionately represented in the justice and prison system — and consult with groups affected by their equations.”
Radical Commentary: With the backdrop of congressional antitrust hearings on Google, Facebook, Amazon, and Microsoft last week, we note that regulation is a natural response to the adoption of new technologies. Textile manufacturing during the industrial revolution in Great Britain, the railway monopolies, and the rise of the US oil industry are some examples of industries that caught regulatory attention as they grew.
Technological revolutions lead to economic and social changes that, in turn, force our institutions to evolve. These are often concurrent waves of change. Today, we are witnessing institutional and governance frameworks that matured in the pre-internet era start to assess and address the centralized power that a handful of companies have developed over the past twenty years.
New Zealand’s is one of a number of governmental policy efforts to ensure that AI, the current technological revolution that is reshaping our lives, is adopted with transparency in our governance institutions.
3) AI and Robotics: How BMW Used Pandemic Plant Stoppages to Boost Artificial Intelligence (Wall Street Journal)
“Manufacturers like BMW that are investing in data analytics, AI and advanced robotics are poised to get a jump on competitors as global markets reopen, investors and industry analysts say. The crisis has acted as a proof-of-concept trial run for quality-control bots and other automated tools in development, said Paul Miller, principal analyst at technology research firm Forrester Research.”
Radical Commentary: At BMW, AI systems were previously checked during planned maintenance shutdowns, often leaving little time to deploy and test large systems. COVID-forced factory shutdowns have created windows of opportunity for faster and larger deployments of automation within manufacturing settings, accelerating adoption and helping companies such as BMW become more resilient to future shocks.
One of our portfolio companies, Covariant, is benefiting from this long term trend. Covariant’s technology advances existing and newly deployed robotic automation by using deep imitation learning, deep reinforcement learning and meta-learning to allow industrial robots to see, reason, and act on the world around them. We expect this trend toward AI-powered automation will continue to accelerate.
4) Ethical Applications of AI: DeepMind and Oxford University researchers on how to ‘decolonize’ AI (Engadget)
“In a moment where society is collectively reckoning with just how deep the roots of racism reach, a new paper from researchers at DeepMind — the AI lab and sister company to Google — and the University of Oxford presents a vision to “decolonize” artificial intelligence. The aim is to keep society’s ugly prejudices from being reproduced and amplified by today’s powerful machine learning systems.
…The “tactics” the paper lists to do this span algorithmic fairness techniques to hiring practices to AI policymaking. It speaks of technologists learning from oppressed communities — giving examples of grassroots organizations like Data for Black Lives — to reverse the colonial mentality of “technological benevolence and paternalism”…
…the authors are calling for a shift away from a longstanding tech culture of supposed neutrality: the idea that the computer scientist just makes tools and is not responsible for their use…
…AI supercharges the idea that those who can’t remember the past are condemned to repeat it: if AI doesn’t remember the past, it will reify, amplify, and normalize inequalities.”
Radical Commentary: There is an important research field focused on thinking about racial inequality within artificial intelligence. The paper referenced in this article draws from research on decolonization studies and argues for better alignment of research and technology development with ethical principles, centering on vulnerable peoples who bear the brunt of negative impacts of innovation and scientific progress.
At Radical we advocate for ethical AI research and applications. Adhering to such values is a requirement in all term sheets when we make new investments.
5) AI Enablers: IBM completes successful field trials on Fully Homomorphic Encryption (Ars Technica)
“FHE is a type of encryption that allows direct mathematical operations on the encrypted data. Upon decryption, the results will be correct. For example, you might encrypt 2, 3, and 7 and send the three encrypted values to a third party. If you then ask the third party to add the first and second values, then multiply the result by the third value and return the result to you, you can then decrypt that result — and get 35.
You don’t ever have to share a key with the third party doing the computation; the data remains encrypted with a key the third party never received. So, while the third party performed the operations you asked it to, it never knew the values of either the inputs or the output. You can also ask the third party to perform mathematical or logical operations of the encrypted data with non-encrypted data”
Radical Commentary: Fully Homomorphic Encryption (FHE) is a complementary technology that enables trust and safety when sharing data within and across organizations. While still early in its development, this technology will have significant implications for industry. For example, it will allow two pharma companies to collaborate on research without direct visibility of each other’s data.
While the compute and memory cost is significantly higher than for non-FHE calculations, we expect early adoption in industries with high regulatory barriers around privacy and where the benefits will outweigh the costs, such as in banking and healthcare.
6) Healthcare Imaging and AI: Digital Health: Enabling the post-COVID-19 transition in imaging (HealthcareIT News)
“The coronavirus pandemic has created a paradigm shift in imaging. Not only have imaging centers faced the need for rapid implementation of strict protocols for patient management, decontamination of equipment and social distancing, these centers have also experienced steep declines in the number of studies being performed…
…To ensure imaging centers are well-positioned for the coming transition and well beyond, digital solutions with artificial intelligence are a must-have. AI-powered digital solutions can aid imaging centers in managing workload via automation, enabling image interpretation and improving efficiency. These solutions can seamlessly integrate into the clinical workflow to alleviate the burden of repetitive tasks and amount of correction steps, which in turn help the radiologists improve their diagnostic accuracy.”
Radical Commentary: The pandemic heavily impacted diagnostic imaging centers and, consequently, patients in need of critical imaging. Those centers that relied on physical CD distribution for sharing patient imagery were helpless when healthcare facilities stopped all non-essential visits. For imaging centers willing to embrace new technologies, the pandemic provides an opportunity to overhaul dated systems and delivery methods and improve the standard of care delivered. Radical’s portfolio company, PocketHealth, a medical image access and sharing platform for patients, doctors and providers (hospitals and clinics), saw an immediate increase in deployments in the wake of COVID. We are seeing this move towards the digitization of legacy analog systems across the healthcare industry. Ultimately this new digital infrastructure will enable wide-scale deployment of AI systems to help deliver personalized and predictive healthcare.
Editor’s Note: We will continue to use this platform to share without commentary articles focused on data and the use of it to illustrate and illuminate racial injustice. Because you cannot fix problems you cannot see or understand.
“Black people and other people of colour make up 83 percent of reported COVID-19 cases while only making up half of Toronto’s population, according to the latest data from the city.
Dr. Eileen de Villa, along with Toronto Mayor John Tory, and other officials, presented new findings and data trends on Thursday afternoon that show racialized and lower-income communities in the city are disproportionately affected by the novel coronavirus.
“Unfortunately, [COVID-19] has had a greater impact on those in our community who face greater health inequities,” said de Villa, who is the city’s medical officer of health.
De Villa said the data shows that Black people account for 21 percent of reported cases in the city while making up only nine percent of the overall population.
Similarly, Arab, Middle Eastern and West Asian people represent 11 percent of the city’s COVID-19 cases, while only making up 4 percent of the total population, de Villa said.
The data also shows that East Asian and white people are underrepresented compared to the size of those populations, de Villa said.”
— R —