Brian Fisher

Universal access is an aim of a potential Digital Canadian Dollar (#CBDC) with inclusivity of people with cognitive disabilities. In this vein the authors involve people with cognitive disabilities in the development of a framework and subsequent guidelines for designers. Importantly this excellent work is applicable to any digital payments and user experience designers can utilize it towards better inclusion. From my colleagues Sebastian Hernandez, Alexandra Sutton-Lalani, John Miedema, Badr Omrane at Bank of Canada in collaboration with Virginie Cobigo, Fatoumata Bah, Munazza Tahir, Danika Lévesque at University of Ottawa / Open Collaboration for Cognitive Accessibility. https://lnkd.in/eVdsqw4w

70

Quantum BC's Ecosystem Start the year off right by checking out local quantum ecosystems to gain insights into research, training, innovation and events like those at Quantum BC (https://quantum-bc.ca/) I've written about many #quantumecosystems from around the world over the years and Canada is replete with quantum environments leading the way in these advanced technology areas. British Columbia, located on the west coast of Canada is the home of Quantum BC. Quantum BC is a "joint initiative led by British Columbia’s top three #research universities that aims to stimulate and enrich collaborative efforts across research, training and innovation in #quantumcomputing. We offer a unique training experience for graduate students in quantum computing hardware and software, to contribute to an exciting and fast-growing industry via internships with our local, national, and international industry partners." Quantum Algorithms Institute Paola Baca Lisa Lambert TRIUMF UBC Blusson Quantum Matter Institute (Blusson QMI) Simon Fraser University The University of British Columbia University of Victoria D-Wave 1QBit #quantum #quantumtechnology #business #technology Image source: Quantum BC

57

1 Comment

Amplifying this new technical guide from Accessibility Standards Canada on accessible and equitable #AI developed by a technical committee chaired by Jutta Treviranus of the Inclusive Design Research Centre at OCAD University. This valuable technical guide focuses on areas where people with #disabilities may face barriers to accessible participation in the design, development, and use of AI systems and equitable treatment by AI systems. Equity refers to fairness, justice, and freedom from discrimination. Equity recognizes that each person has different circumstances and focuses on enabling all individuals to achieve the same outcomes. People with disabilities experience the extremes of both the opportunities and risks brought about by the deployment of AI. This guide offers guidance and directives that support ethical practices when implementing AI, which includes providing proactive guidance to prevent harm for people with disabilities in emerging applications of AI. For instance, the guide recommends that organizations: • Recognize that people with disabilities should be able to participate fully as stakeholders in all roles in the AI lifecycle. • Recognize that AI systems need to be accessible to people with disabilities. • Design, develop, procure, and/or customize AI systems that are accessible and equitable. • Conduct ongoing impact assessments, ethics oversight and monitoring of potential harms • Train personnel in accessible and equitable AI • Provide transparency, accountability and consent mechanism • Provide access to equivalent alternative approaches • Provide feedback, complaints, redress and appeals mechanisms ... and more! Check out the full technical guide here: https://lnkd.in/eV_raH7H #accessibleAI #a11y #accessibility #equitableAI #deia #responsibleAI #inclusiveAi #disabilityInclusion #diversityEquityInclusion

54

3 Comments

Fascinating read on how alignment/misalignment, especially in language and cultural aspects lead to potential & unfortunate (unintended) biases in LLMs https://buff.ly/3Ax5XQ3 #ai #llm #bias #humancenteredAI

CALL FOR PROPOSALS: The rapid development of AI, particularly generative AI, relies on millions of workers engaged in data enrichment – curating, labeling, and annotating data to train AI models. Many of these workers often work in precarious contract jobs or pay-per-task environments in the Global South and are virtually invisible in the global economy. As a result, their contributions to the global AI industry and what measures are in place to protect them are not yet well understood. That’s where you come in! To address this critical information gap, USAID is currently seeking a qualified research team to explore the data enrichment labor market in the Global South and its role in global AI value chains. This research will provide invaluable insights into this historically undervalued workforce and identify potential approaches to support and protect workers in the AI era. Learn more and apply by January 31, 2025: https://lnkd.in/epNf4mFu

4

Last week I gave a talk titled "Search and Society: Reimagining Information Access for Radical Futures" at the Canadian AI Responsible AI event. Slides: https://lnkd.in/estEC4QV If you download the slides, the speaker notes contain the full transcript of my talk. This talk touched on 3 of my recent papers/manuscripts. 1. I talked about the Consequences-Mechanisms-Risks (CMR) framework introduced in our #FAccT2024 paper. https://lnkd.in/ehvFpvAW 2. Then I discussed the sociotechnical implications of generative AI for information access. This was based on an early draft of our chapter for a book that Chirag Shah and Ryen White are co-editing. https://lnkd.in/eVGbsSnt 3. Finally, I talked about challenging today's homogenized sociotechnical imaginaries that drive information retrieval research in favor of diverse visions of possible futures informed by prefigurative politics and centering IR research on societal needs. https://lnkd.in/ehqjg3GQ It was my first time attending the Canadian AI conference and the Responsible AI event. I thoroughly enjoyed all the conversations. Big thanks to Ebrahim Bagheri for inviting me to give a talk at the event.

38

Many comments from Geoff Hinton and myself in this nice CBC article. Do you disagree with us on any of them? My takes: "The stakes are extremely high. If we get this right, we get tremendous upside. And if we get this wrong, we might be talking about the end of human civilization." "I definitely think that there's a chance, and a non-trivial chance, that [superintelligence] could show up this year," he said. "We have entered the era in which superintelligence is possible with each passing month and that probability will grow with each passing month." Clune estimates there could be as high as a 30 to 35 per cent chance that everything goes extremely well in terms of humans maintaining control over superintelligences, meaning areas like health care and education could improve beyond our wildest imaginations. He also says superintelligence would help us "make death optional" by turbo-charging science and eliminating everything from accidental death to cancer. "Since the dawn of the scientific revolution, human scientific ingenuity has been bottlenecked by time and resources," he said. "And if you have something way smarter than us that you can create trillions of copies of in a supercomputer, then you're talking about the rate of scientific innovation absolutely being catalyzed." "What worries me a lot less than the companies developing it are the other countries trying to catch up and the other organizations that have far less scruples than I think the leading AI labs do." One practical solution Clune offers, similar to the nuclear era, is to invite all of the major AI players into regular talks. He believes everyone working on this technology should collaborate to ensure it's developed safely. "This is the biggest roll of the dice that humans have made in history and even larger than the creation of nuclear weapons," Clune said, suggesting that if researchers around the world keep each other abreast of their progress, they can slow down if they need to. More in the article, especially Geoff's takes. https://lnkd.in/gG9cKAw4

18

4 Comments

As the generation and application of #RWE continues to evolve, the use of #ExternalControlArms (ECAs) constructed from #RWD has emerged as a key strategy to strengthen regulatory decision-making, and this has been further enhanced by rapid advancements in #AI. Here we speak with Paul Arora (University of Toronto) about the key guidelines available on the use of ECAs, the common challenges and opportunities in utilizing RWD for ECAs and emerging role of generative AI. Read at The Evidence Base® #outcomesresearch #realworldevidence #realworlddata #regulation #regulatoryaffairs #healthdata #healthequity #healthpolicy #artificialintelligence

40

3 Comments

There was such a great discussion about public input and user research based on my post last week that added a few thoughts, pulling on threads from Peter Bonner, 🚀 Molly Cain MBA, CVP, Marina Nitze and quoting Genevieve Gaudet. This one talks not just who we ask for input, but when, and what we for input on. And comes back to the connection between complexity and capture, and the role of product management in countering capture. https://lnkd.in/gjaePqh9

70

2 Comments

Today’s Nobel Prize in Physics Congratulations to Geoffrey Hinton for being a co-recipient of the 2024 Nobel Prize in physics for his work in Machine Learning. Some articles this morning have coined him as the grandfather of AI.   Dr. Hinton is a favourite topic of Canadian IP pundits. By all accounts his AI work at UoT was groundbreaking and, along the way, patents associated with some of it are now owned by Google. It is here that the pundits entered the picture with chatter about the loss of Canadian IP.   I too entered the chatter in 2020, sort of. Okay, it was more like talking to myself. At that time I penned a series of three articles about IP in Canada. The first or these (https://lnkd.in/esE7eKc) considered the loss of technology and IP. I said we need to identify the technology that is being developed and patented and we need to invest.  We need to be on the same playing field as those buying Canadian IP. This is one of my motivations for founding IPmart.   A September 15th Opinion in the Globe & Mail framed the IP discussion in terms of productivity. The authors wrote “… we need to dramatically increase the rate at which we protect and commercialize new ideas and the intellectual property (IP) that often starts in publicly funded research.” Hmmm … that sounds familiar! #technology #patents #CDNtech #IPmart

8

Check out the first story from 🇨🇦 "Canadian Achievements in Science and Technology Week: "University of Victoria’s AI Revolutionizes Space Debris Tracking" to discover how Canadian scientists use machine learning to prevent satellite collisions and secure the skies above us! #canada #canadian #science #technology #AI #space

1

A few questions for those who conduct #Research with Internal users/Employees (particularly in a #B2B environment). 🔹 What is your current process for capturing Informed Consent from internal users/employees? 🔹 What information do you provide internal users/employees prior to taking part in research as part of the consent process? 🔹 How do you incentivise internal users/employees who choose to take part in research? Appreciate any shares/signal boosts to these questions within your own networks, thanks.

Related to this question; The Department for Education UR Team provide a great example of #WorkingInTheOpen with the content they publish on ‘Research with Internal Users’. https://lnkd.in/e5_DksFT Have you seen any other examples of published content looking at this area? Thanks.

1

Are you concerned about the potential impact of AI assistants on the spread of misinformation? Our own Josh Goldstein recently co-authored a report chapter, in partnership with researchers at Google, that explores this issue in depth. They highlight four key risks: 1️⃣ Increased vulnerability to misinformation if users develop uncritical trust in AI assistants. 2️⃣ Entrenched viewpoints and reduced political efficacy if AI assistants personalize biased or partial content to match users’ pre-existing views. 3️⃣ Degraded and homogenized information environments if AI assistants contribute to the proliferation of low-quality content. 4️⃣ Weaponized influence agents if propagandists manipulate or hack widely used AI assistants. But don't worry, the chapter also offers solutions! It proposes a combination of technical approaches (like restricting specific applications and developing better detection mechanisms) and policy solutions (like transparency requirements and public education). Learn more by reading Chapter 16 or check out the full paper here: https://lnkd.in/eA6gTdQg

30

Explainable AI (XAI) decisions found to be biased towards WEIRD societies that favour individualistic over interdependent explanations. A new meta analysis has reviewed 200 studies on XAI outputs and found 94% did not indicate awareness of cultural variations in outputs. Only 52% of the studies reported on cultural background. Of these 81% only sampled western, industrialised, educated, rich and democratic societies (WEIRD). A mere 8.4% sampled non-western populations and 10.3% sampled mixed populations. This matters for designing inclusive systems of explainable AI. There are two common ways to explain someone’s actions. One involves invoking the person’s beliefs and desires. This explanation is internalist, focused on what’s going on inside someone’s head. The other is externalist, citing factors like social norms, rules, or other factors that are outside the person. Many psychological studies suggest internalist explanations are preferred in “individualistic” countries where people often view themselves as more independent from others. These countries tend to be in the west, educated, industrialised, rich, and democratic. However, such explanations are not obviously preferred over externalist explanations in “collectivist” societies, such as those commonly found across Africa or south Asia, where people often view themselves as interdependent. Preferences in explaining behaviour are relevant for what a successful XAI output could be. An AI that offers a medical diagnosis might be accompanied by an explanation such as: “Since your symptoms are fever, sore throat and headache, the classifier thinks you have flu.” This is internalist because the explanation invokes an “internal” state of the AI – what it “thinks” – albeit metaphorically. Alternatively, the diagnosis could be accompanied by an explanation that does not mention an internal state, such as: “Since your symptoms are fever, sore throat and headache, based on its training on diagnostic inclusion criteria, the classifier produces the output that you have flu.” This is externalist. The explanation draws on “external” factors like inclusion criteria, similar to how we might explain stopping at a traffic light by appealing to the rules of the road. Why the results matter: If findings about explainable AI systems only hold for one kind of population, these systems may not meet the explanatory requirements of other people affected by or using them. This can diminish trust in AI. When AI systems make high-stakes decisions but don’t give a satisfactory explanation, they’ll likely be distrusted even if their decisions (such as medical diagnoses) are accurate. Summary adapted from source: https://lnkd.in/enjTuHXz

11