We have been proudly investing the Canada Pension Plan Fund for 25 years, helping ensure that more than 22 million Canadians have financial security in retirement.
As Chief Operating Officer at CPP Investments, Jon Webster is on the front lines of the technological change reshaping global investing. In the latest episode of the Technovation podcast with Peter High, he reflects on the importance of connecting technology to value creation, the “heart of data security”, and how generative AI will reconfigure work.
**The following has been edited for clarity and brevity.
Peter High: The past year has coincided with remarkable change in the ability to process data, with further advances in artificial intelligence, generative artificial intelligence, and the like. How has this impacted the work your team is doing?
Jon Webster: One of the foundational things we’ve done is to make sure we can connect technology, frankly, to value. While data is really important, that’s not the only thing that goes into the mix in terms of creating value. So one of the shifts we’ve been making is to connect technology to the product owners and the teams that can create value with it.
Once you’re connecting people to the technology that they can do something valuable with, and you’ve got a modular, responsive, nimble architecture, then you can start to pull data from different systems, different platforms, liberate the data and start to do, I think, interesting value creation things with it.
Generative AI has been a real step change in that ability to democratize technology over the last 12 months. I’m actually very bullish on it. I think generative AI large language models unlock capabilities that allow us to change the nature of our interactions, the nature of how we interact with intelligence. We’ve already got very broad access to generative AI capabilities across the organization and have very targeted select value creation opportunities including in investment strategy.
Peter High: Talk a bit about how you organize your team, and how that team, a global one, collaborates in a way that makes it effective.
Jon Webster: I think there’s two parts to being effective in a global organization with technology. You need to focus on capabilities that can be leveraged enterprise-wide. Having enterprise-wide CRM systems, pipeline systems, making sure we’re understanding all the interactions around our partners and how deals move through their life cycles. That consistent, integrated view, I think is really important.
But in a global organization, you also really need to connect technology, frankly, to the everyday user. If you sit and only think about the enterprise capabilities, you miss real opportunities to connect people to everyday value-creation opportunities.
Peter High: Clearly, by being global, and being able to sit with the intended users of the technology, you gain special insights as to their uses. What are some of the things that you are learning from those deeper collaborations?
Jon Webster: Generative AI is a really good example. We’ve made a deliberate decision to start with a very hands-on experience in one of our global offices, to really concentrate on getting every user two or three hours of hands-on facilitated experience with tools like Microsoft Copilot and our own internal GPT-4 based platforms.
What we’re finding with generative AI, at the moment, is it can do some stuff. Then maybe it lets you down on some stuff, and then maybe it exceeds your expectations on some other things. We’re finding that colleagues who really stick with using generative AI-based tools need to pass a two or three-hour bar of immersive usage with it before they get comfortable to say, “Actually, I’m fine to discover what works and what doesn’t work.”
When we’ve seen colleagues engage for 25, 30 minutes and it doesn’t quite meet their needs, it’s a little bit disappointing off the bat. There’s a danger of giving up on generative AI. Actually, we’re big fans of Ethan Mollick here. If you’re familiar with Ethan Mollick, who talks a lot about generative AI on his podcast, he says you need 10 hours with a GPT-4 class model before you really start to believe that generative AI can do something for you.
Peter High: I wonder if you could talk a little bit about the way in which you manage security given the remarkable amount of data that flows through your systems?
Jon Webster: I think the heart of good security is getting people to take it seriously. We’re very fortunate that we’ve really been able to get our product teams, our product managers to take a whole product perspective. That is, not just think about new features, but the resilience of the platforms they’re building, the data that’s within them, the entitlements of that data, how that data is made consumable, accessible across the organization.
The second piece, which is really important to us, and frankly, is a real advantage, is we’re largely software defined. We made the move to the cloud in 2019 and we have very little on-premise left. Of course, we have SaaS solutions, but the majority of our built software is all built on public cloud. Being software-defined allows us to be very effective in our secure by design principles. We’re able to embed security checking, vulnerability checking, vulnerability scanning with our pipelines to make sure that our software development is sound and secure from the outset.
The third piece, which is super important, is to just have that in a much broader context, where you’re thinking about the full life cycle of security.
Now we’re fortunate that we’ve got deep intelligence expertise within the organization. Everything from our threat intelligence, how we think about the world, the threats we need to consider, all the way through to the expertise in our security operations. In all of that software development, that whole-product thinking, security is the first thing embedded. Now, you can never take anything for granted with security, so you’ve got to continue to up your game. The world gets more complicated. Threat actors get more ambitious, so we don’t take anything for granted.
Peter High: We’ve certainly talked about important trends, perhaps chief among those, artificial intelligence and Gen AI. Any other trends as you look to the future that have you particularly excited, Jon?
Jon Webster: Actually, I’m genuinely very excited by generative AI. I know we’ve talked about it already, but I do think the possibilities of growing very different technology experiences with generative AI are starting to appear. One of the thought experiments I do with the team here, and we’re going to hopefully start some work on this soon, is to confine yourself to ChatGPT 4, confine yourself to only using AI agents. How would you change the way you do your work today?
I think that’s a fascinating experiment that almost every organization can run in almost in every part of the work that they do. I really do see work reconfiguring. When I look at what we do in most organizations, a lot of the structure we have is driven from two things. One is trying to bring people together at a point in time to collaborate. You need to bring tax expertise in, investment expertise in, legal expertise in. A lot of the structure is about choreographing, getting all that expertise together at the same point in time. Then the other part of that structure is about evidence and control, that you’ve made some meaningful step along the way.
A lot of our structure is just to facilitate great conversations. Why I’m so excited about generative AI is, I think it just gives us a different way of doing that. I love the idea of just starting to grow new and different experiences of interacting with intelligence.
{:en}
Lenin’s well-known quote “There are decades where nothing happens; and there are weeks where decades happen” could very well be applied to the field of artificial intelligence.
In November 2022, artificial intelligence (AI) went mainstream. OpenAI, the San Francisco research lab backed by Microsoft to the tune of US$13 billion, unleashed ChatGPT, an application capable of writing poetry, answering questions, generating code, and planning vacations, among other pursuits. Two months later, Alphabet followed with its own LaMDA-powered experimental chatbot, Google Bard, and the race was on.
The Age of AI has begun, ushered in by the transformative power of generative AI, a type of artificial intelligence that can “generate” or produce new content, including text, images, music, video, synthetic data, and more when directed with simple text prompts in natural human language. We’re still in the early days of this technology, but it’s increasingly clear that generative AI represents a transformative technology that will change the way we learn, work, and create across all industries, with massive long-term economic and societal implications.
The combination of a natural language user interface and massive online digital distribution will increase access to this technology far beyond its usual developer-community confines, touching everyone with an internet connection. We expect this new era in AI to have long-term investment and operational implications for investors.
Source: State of the Developer Nation Report, Gartner, World Bank and CPP Investments analysis
A conceptual framework to research investment opportunities in AI
We believe AI is the next platform shift after personal computing, mobile, and the cloud. At CPP Investments, we’re using the following framework to help determine potential investment opportunities:
Source: CPP Investments analysis
This framework continues to evolve as many moving parts impact an investor’s ability to forecast how this value chain will develop with certainty. Investors, like us, will need to iterate their conceptual models frequently, and challenge assumptions on where value will accrue.
We anticipate companies across all industries will be compelled to develop their own internal AI roadmaps, experiments, and use cases for deploying generative AI applications. This will likely lead to a series of new investment opportunities as companies deploy strategies and strike up partnerships across the AI and cloud computing eco-systems. We’re seeing this dynamic already starting to take shape within the technology eco-system, with partnerships forged between Microsoft and OpenAI, Google and Anthropic, and Amazon and Hugging Face, as early examples.
A growing number of software as a service (SaaS) companies may redefine their product roadmaps and demonstrate how they plan to incorporate generative AI into existing applications and new products. There may also be a growing role for cybersecurity companies in the generative AI era, to prevent bad actors and incorporate more advanced data protection.
On the technical front, we’ll be monitoring how large language models (LLMs) will continue to evolve in size, reliability, ability to predict with even higher levels of accuracy, and understanding of human context. We will also monitor how LLMs grow, over the long-term, in their ability to reason through human interaction with natural language.
The importance of investing responsibly in AI
The AI revolution is profound but it’s similar to earlier revolutions (pre-industrial, industrial, internet, etc.) that have led to structural and societal change – although opinions abound about the speed and magnitude of AI’s impact. While the industrial revolution catalyzed social change (e.g., the growth of cities and rise of the middle class), the underpinning – and continued consumption – of traditional energy has also given us climate change, one of the most pressing challenges facing society today. Can we learn from history to be more pro-active in managing the long-term impacts of the AI revolution?
Even as AI generates consumer benefits and business value, it also gives rise to a host of risks. Disparate outcomes such as biased hiring, unfair lending practices, incorrect medical diagnoses, and runaway surveillance expose businesses to reputational damage, political risk, legal action, and loss of consumer trust. Investors will likely need to change the way they diligence and invest in increasingly AI-enabled businesses, as well as develop governance frameworks on how portfolio companies adopt AI-enabled tools for their operations.
Organizations around the world, from the Organization for Economic Co-operation and Development to big tech firms, are working to develop principles around responsible artificial intelligence, or RAI (an approach to developing, assessing, and deploying AI systems in a safe, trustworthy, and ethical way) to minimize risks and optimize outcomes. Together, the CPP Investments Insights Institute and the World Economic Forum have spoken with over 80 experts, from institutional investors and academics to representatives from venture capital firms and technology companies, as well as other sectors of the AI ecosystem, to explore what it means to invest responsibly in artificial intelligence.
These interactions with experts have exposed both the immense potential of AI and the material benefits of investing in it responsibly. The discussions have also highlighted the need for appropriate metrics to assess AI’s alignment with and performance against RAI principles, and for practical methodologies to integrate AI-related factors into investment decisions.
Over time, investors will likely need to adopt similar principles and frameworks for other emerging business risks, like climate change. If it is a material opportunity for a company to deploy AI into its operations, we expect the company board to ensure it is appropriately overseeing and counseling the executive team to do so in such a way where risks are mitigated, and opportunities are appropriately captured. To this end, CPP Investments expanded our definition of sustainability-related factors to include responsible sourcing and use of AI in our recently updated 2023 Policy on Sustainable Investing.
Looking ahead
AI will be an important part of businesses everywhere regardless of sector or geography. Given the significant impact AI will undoubtedly have, investors will need to consider the long-term implications of AI and the associated risks and opportunities, and to determine effective ways to use AI to preserve and create enduring value.
Authors
Nadeem Janmohamed
Managing Director, Active Equities North America
Nadeem Janmohamed, Managing Director in our Active Equities North America team, leads CPP Investments’ Technology investment domain, which spans public and pre-IPO equity investments in software, cloud, semiconductors, IT hardware and electronic systems.
Chris Gillam
Director, Operational Risk
Chris Gillam, a Director in our Operational Risk team, is on secondment to the CPP Investments Insights Institute where he is studying technology disruption. Chris is also a Fellow with the World Economic Forum’s Responsible Investing in Artificial Intelligence project.
Our Expertise
“Advancing a Thriving World” – Views from the 25th annual Milken Institute
The questions asked, and themes raised, promise to dominate headlines in the years to come.
Article
•
June 9, 2023
Technology Disruption
Views on Artificial Intelligence (AI) with Jordan Jacobs, Co-Founder &
As AI changes the future of work, what’s the most important skill that schools need to teach? Hear Jordan’s take.
Video
•
June 9, 2023
Our Expertise
Women at the forefront of alternative investments
This Q&A was first published by iConnections, what follows are insights from CPP Investments leaders.
Article
•
June 7, 2023
{:}{:fr}
La célèbre citation de Lénine, « Il y a des décennies où rien ne se passe et des semaines où des décennies se produisent », pourrait très bien s’appliquer au domaine de l’intelligence artificielle.
En novembre 2022, l’intelligence artificielle (IA) est devenue monnaie courante. OpenAI, le laboratoire de recherche de San Francisco que Microsoft a soutenu à hauteur de 13 milliards de dollars américains, a lancé ChatGPT, une application capable d’écrire de la poésie, de répondre à des questions, de générer des codes et de planifier des vacances, entre autres. Deux mois plus tard, Alphabet a suivi avec son propre dialogueur expérimental alimenté par LaMDA, Google Bard, et la course était lancée.
L’ère de l’IA a commencé, marquée par le pouvoir transformateur de l’IA générative, un type d’intelligence artificielle qui peut « générer » ou produire du nouveau contenu, notamment du texte, des images, de la musique, des vidéos, des données synthétiques et plus encore, lorsqu’une requête est effectuée au moyen de messages texte simples dans un langage humain naturel. Nous n’en sommes qu’au début de cette technologie, mais il est de plus en plus évident que l’IA générative représente une technologie transformatrice qui changera notre façon d’apprendre, de travailler et d’innover dans tous les secteurs, avec des implications économiques et sociétales massives à long terme.
La combinaison d’une interface utilisateur en langage naturel et d’une distribution numérique massive en ligne élargira l’accès à cette technologie bien au-delà des limites habituelles de la communauté des développeurs, en touchant tous ceux qui disposent d’une connexion à l’Internet. Nous pensons que cette nouvelle ère de l’IA aura des implications opérationnelles et en matière de placement à long terme pour les investisseurs.
Sources : Rapport State of the Developer Nation 2022, Gartner, Banque mondiale et analyse d’Investments RPC.
Un cadre conceptuel pour la recherche d’occasions de placement dans l’IA
Nous pensons que l’IA est la prochaine évolution en matière de plateforme après l’informatique domestique, mobile et nuagique. À Investissements RPC, nous utilisons le cadre suivant pour déterminer les occasions de placement potentielles :
Source : Analyse d’Investissements RPC.
Ce cadre continue d’évoluer, car de nombreux éléments en évolution ont une incidence sur la capacité d’un investisseur à prévoir avec certitude la façon dont cette chaîne de valeur évoluera. Les investisseurs, comme nous, devront fréquemment répéter leurs modèles conceptuels et remettre en question les hypothèses quant à savoir où la valeur s’accroîtra.
Nous prévoyons que les entreprises de tous les secteurs seront obligées d’élaborer leurs propres feuilles de route, expériences et cas d’utilisation d’IA pour le déploiement d’applications d’IA générative. Il en résultera probablement une série de nouvelles occasions de placement à mesure que les sociétés déploieront des stratégies et établiront des partenariats dans les écosystèmes de l’IA et de l’infonuagique. Cette dynamique commence déjà à prendre forme au sein de l’écosystème technologique, comme en témoignent, par exemple, les partenariats établis entre Microsoft et OpenAI, Google et Anthropic, Amazon et Hugging Face.
Un nombre croissant de sociétés qui fournissent des logiciels-services (SaaS) peuvent redéfinir leur feuille de route de produits et montrer comment elles prévoient intégrer l’IA générative aux applications existantes et aux nouveaux produits. Les entreprises de cybersécurité pourraient également jouer un rôle de plus en plus important à l’ère de l’IA générative, afin de prévenir les mauvais acteurs et d’intégrer une protection des données plus avancée.
Sur le plan technique, nous surveillerons la façon dont les grands modèles de langage continueront d’évoluer sur le plan de la taille, de la fiabilité, de la capacité à prédire avec encore plus d’exactitude le langage humain et à le comprendre. Nous surveillerons également l’évolution à long terme de la capacité des grands modèles de langage à raisonner par l’interaction humaine avec le langage naturel.
L’importance d’investir de façon responsable dans l’IA
La révolution de l’IA est profonde, mais elle ressemble aux révolutions antérieures (préindustrielle, industrielle, Internet, etc.) qui ont conduit à des changements structurels et sociétaux, bien que les opinions abondent sur la vitesse et l’ampleur de l’impact de l’IA. Si la révolution industrielle a catalysé le changement social (p. ex., la croissance des villes et l’essor de la classe moyenne), la consommation sous-jacente et continue de l’énergie traditionnelle nous a également apporté le changement climatique, l’un des défis les plus pressants auxquels la société est confrontée aujourd’hui. Pouvons-nous tirer des leçons de l’histoire pour être plus proactifs dans la gestion des répercussions à long terme de la révolution de l’IA?
Même si l’IA génère des avantages pour les consommateurs et de la valeur pour les entreprises, elle engendre également une série de risques. Des résultats contradictoires, comme des embauches biaisées, des pratiques de prêt déloyales, des diagnostics médicaux erronés et une surveillance incontrôlée, exposent les entreprises à des atteintes à leur réputation, à des risques politiques, à des poursuites judiciaires et à la perte de confiance des consommateurs. Les investisseurs devront probablement changer leur façon de diligenter et d’investir dans des entreprises de plus en plus axées sur l’IA, ainsi que développer des cadres de gouvernance sur la façon dont les entreprises en portefeuille adoptent des outils axés sur l’IA pour leurs opérations.
Des organisations du monde entier, qu’il s’agisse de l’Organisation de coopération et de développement économiques ou des grandes sociétés de technologie, travaillent à l’élaboration de principes en matière d’intelligence artificielle responsable, ou IAR (une approche de développement, d’évaluation et de déploiement des systèmes d’IA d’une manière sûre, digne de confiance et éthique), pour réduire les risques au minimum et optimiser les résultats. Ensemble, l’Institut sur les données d’Investissements RPC et le Forum économique mondial se sont entretenus avec plus de 80 experts, qu’il s’agisse d’investisseurs institutionnels et universitaires ou de représentants de sociétés de capital de risque et de sociétés de technologie, ainsi que d’autres secteurs de l’écosystème de l’IA, afin d’explorer ce que signifie investir de façon responsable dans l’intelligence artificielle.
Ces interactions avec des experts ont mis en lumière l’immense potentiel de l’IA et les avantages importants d’y investir de façon responsable. Les discussions ont également souligné la nécessité d’avoir des mesures appropriées pour évaluer l’harmonisation de l’IA avec les principes de l’IAR et son rendement par rapport à ceux-ci, ainsi que des méthodologies pratiques pour intégrer les facteurs liés à l’IA dans les décisions de placement.
Au fil du temps, les investisseurs devront probablement adopter des principes et des cadres similaires pour d’autres risques d’affaires émergents, comme le changement climatique. Si une entreprise a la possibilité de déployer l’IA dans ses activités, nous nous attendons à ce que le conseil d’administration supervise et conseille l’équipe de direction de manière à atténuer les risques et à saisir les occasions convenablement. À cette fin, Investissements RPC a élargi sa définition des facteurs liés à la durabilité pour y inclure l’approvisionnement responsable et l’utilisation de l’IA dans sa Politique en matière d’investissement durable 2023 récemment mise à jour.
Perspectives
L’IA occupera une place importante dans les entreprises partout, peu importe le secteur ou la région. Compte tenu de l’impact considérable que l’IA ne manquera pas d’avoir, les investisseurs devront examiner les implications à long terme de l’IA, les risques et les occasions qui y sont associés, et déterminer les moyens efficaces de l’utiliser pour préserver et créer une valeur durable.
Auteurs
Nadeem Janmohamed
Directeur général de notre équipe Actions à gestion active en Amérique du Nord
Nadeem Janmohamed, directeur général de notre équipe Actions à gestion active en Amérique du Nord, dirige le secteur des placements technologiques d’Investissements RPC, qui couvre les placements en actions dans des sociétés ouvertes et à des phases de financement précédant un PAPE dans les segments des logiciels, de l’infonuagique, des semi-conducteurs, du matériel informatique et des systèmes électroniques.
Chris Gillam
Directeur de notre équipe Risque opérationnel
Chris Gillam, un directeur de notre équipe Risque opérationnel, est en affectation provisoire à l’Institut sur les données d’Investissements RPC, où il analyse la technologie de rupture. Chris participe également au projet Responsible Investing in Artificial Intelligence du Forum économique mondial.
Our Expertise
“Advancing a Thriving World” – Views from the 25th annual Milken Institute
The questions asked, and themes raised, promise to dominate headlines in the years to come.
Article
•
June 9, 2023
Technology Disruption
Views on Artificial Intelligence (AI) with Jordan Jacobs, Co-Founder &
As AI changes the future of work, what’s the most important skill that schools need to teach? Hear Jordan’s take.
Video
•
June 9, 2023
Our Expertise
Women at the forefront of alternative investments
This Q&A was first published by iConnections, what follows are insights from CPP Investments leaders.
Article
•
June 7, 2023
{:}
