Note: While we are actively monitoring news out of the banking sector and will follow up if new information requires commentary, we feel that recent developments in artificial intelligence are both notable and newsworthy.
Although we all watch the markets on a day-to-day basis, some of the best investment decisions are made with long-term goals. Autos were the trend of the 1950s, airlines of the 1960s, software in the 70s, personal computers in the 80s, and so on until mobile phones and social media of the last couple of decades. Looking forward, is Artificial Intelligence (“AI”) a viable long-term investment theme?
Currently, the hot name in the artificial intelligence space is OpenAI, a private software and platform company founded in 2015 with $1 billion. Some of the original founders were Elon Musk (since retired from the board), other members of the “PayPal mafia” (founders), and Reid Hoffman (co-founder of LinkedIn).
Since OpenAI’s ChatGPT-3 was introduced to the public in November 2022, AI has been the buzzword bandwagon that all tech companies have jumped on. This is interesting because these same companies have been investing billions of dollars into their own AI projects, and most AI spending has gone to behind the scenes projects like more personalized news feeds in Facebook, better ad-tracking to products you might buy on Amazon, or simply better FAQ functions on company’s websites.
ChatGPT introduced generative AI , a conversational way to access data and refine results from the database to the masses. Microsoft, an early investor in OpenAI, has invested another $10 billion into the company and is rolling out generative AI across all of its platforms. Google has announced its response to ChatGPT with Bard which, embarrassingly, returned incorrect data on its public rollout. Google is also using AI in more of its search functions and to better match advertisers to searchers. On March 14, 2023, OpenAI introduced GPT-4 and announced a list of partners that are integrating the new technology into their corporate applications. Prominent names include Stripe, Kahn Academy, the Government of Iceland, and Morgan Stanley.
Some of the service and software companies with the largest inroads into AI so far are Google, Microsoft, Amazon, Facebook, Baidu and Apple. We know, however, that rarely do these behemoths of industry come up with their own cutting edge technology. They find it in smaller companies like OpenAI. Similarly on the hardware side of the equation, large chip companies like NVIDIA, AMD, Xilinx, Intel, and Qualcomm have the capabilities and reach to exploit discoveries in chip technology but might not be the first to discover the technology.
Perhaps a better way to look for the nascent forms of innovation is to look at the types of technologies and find companies making strides in the building blocks of usage. Each type of AI is designed to perform specific tasks and functions.
Reactive Machines: These are the most basic types of AI systems that are designed to respond to specific stimuli, but they don’t have memory or the ability to learn. Examples include Deep Blue, which was designed to play chess, and AlphaGo, which was created to play the boardgame Go.
Limited Memory: These AI systems can use past experiences to make decisions, such as self-driving cars that can recognize objects and make decisions based on their previous interactions with similar objects.
Theory of Mind: These AI systems can understand human emotions, beliefs, and desires, and can use this information to interact with humans in a more natural and intuitive way.
Self-Aware: While we do not currently have self-aware AI, it is a topic of much research and speculation. These would be the most advanced AI systems that have consciousness and the ability to think and reason like humans.
Narrow AI: AI systems that are designed to perform specific tasks or functions, such as speech recognition, image recognition, or natural language processing.
General AI: AI systems that can perform any intellectual task that a human can, and can learn and adapt to new situations in real-time.
One of the most exciting uses for AI is image recognition, matching, and generation. It can be as simple as “paint a portrait of Patrick Mahomes in the style of Leonardo da Vinci” or as complicated as “a photo of a giraffe wearing sunglasses and a hat riding a motorcycle in Paris.”
These are crude examples, but other so-called “deepfakes” of photos and videos are much more sophisticated and designed to deceive. In the near future, we may not be able to trust what we see or hear. This leads to a very interesting rabbit hole of AI, verification.
Currently, our verification methods (DocuSign, two-factor logins, PIN codes) are designed to identify and verify senders and receivers of documents or data. With AI, there is the added step of verifying the authenticity of the data itself. This can be accomplished using blockchain technology, which has also been used to launch thousands of crypto currencies and NFTs (non-fungible tokens). NFTs function as an electronic “receipt” verifying the source of data. Used directedly, blockchain is not simply a way of verifying transactions or proving that you own an animated image of an ape, but a way to authenticate any form of data that you receive.
Although the field of artificial intelligence has been around for years, the rate of change in innovation and application has increased exponentially over the past 6 months and will continue to do so. AI is a subject that is changing so rapidly that much of what we’ve written might be moot within the next year. However, whether we recognize it around us or not, AI is going to become a much larger part of our everyday life than we could have imagined even two years ago.
Although investors are not yet able to directly invest in most pure play companies, like Open AI, we can participate through larger firms who can better deploy these emerging technologies. This is also a time to be wary of anyone tagging “AI” onto their name or mission statement – there are several parallels to be drawn to the similar trend with “crypto” or “blockchain” a few years ago. As always, the best way to play a new technology, or theme, in the market is to be widely diversified in that industry. We feel that we have well-diversified exposure to many components of the AI space through investments in software, hardware, and platform providers, and we look forward to future developments.
Market Commentary: March 2023 – Investing in the World of AI
Investing in the World of AI
March 23, 2023
By Ward Williams
Note: While we are actively monitoring news out of the banking sector and will follow up if new information requires commentary, we feel that recent developments in artificial intelligence are both notable and newsworthy.
Although we all watch the markets on a day-to-day basis, some of the best investment decisions are made with long-term goals. Autos were the trend of the 1950s, airlines of the 1960s, software in the 70s, personal computers in the 80s, and so on until mobile phones and social media of the last couple of decades. Looking forward, is Artificial Intelligence (“AI”) a viable long-term investment theme?
Currently, the hot name in the artificial intelligence space is OpenAI, a private software and platform company founded in 2015 with $1 billion. Some of the original founders were Elon Musk (since retired from the board), other members of the “PayPal mafia” (founders), and Reid Hoffman (co-founder of LinkedIn).
Since OpenAI’s ChatGPT-3 was introduced to the public in November 2022, AI has been the buzzword bandwagon that all tech companies have jumped on. This is interesting because these same companies have been investing billions of dollars into their own AI projects, and most AI spending has gone to behind the scenes projects like more personalized news feeds in Facebook, better ad-tracking to products you might buy on Amazon, or simply better FAQ functions on company’s websites.
ChatGPT introduced generative AI , a conversational way to access data and refine results from the database to the masses. Microsoft, an early investor in OpenAI, has invested another $10 billion into the company and is rolling out generative AI across all of its platforms. Google has announced its response to ChatGPT with Bard which, embarrassingly, returned incorrect data on its public rollout. Google is also using AI in more of its search functions and to better match advertisers to searchers. On March 14, 2023, OpenAI introduced GPT-4 and announced a list of partners that are integrating the new technology into their corporate applications. Prominent names include Stripe, Kahn Academy, the Government of Iceland, and Morgan Stanley.
Some of the service and software companies with the largest inroads into AI so far are Google, Microsoft, Amazon, Facebook, Baidu and Apple. We know, however, that rarely do these behemoths of industry come up with their own cutting edge technology. They find it in smaller companies like OpenAI. Similarly on the hardware side of the equation, large chip companies like NVIDIA, AMD, Xilinx, Intel, and Qualcomm have the capabilities and reach to exploit discoveries in chip technology but might not be the first to discover the technology.
Perhaps a better way to look for the nascent forms of innovation is to look at the types of technologies and find companies making strides in the building blocks of usage. Each type of AI is designed to perform specific tasks and functions.
One of the most exciting uses for AI is image recognition, matching, and generation. It can be as simple as “paint a portrait of Patrick Mahomes in the style of Leonardo da Vinci” or as complicated as “a photo of a giraffe wearing sunglasses and a hat riding a motorcycle in Paris.”
These are crude examples, but other so-called “deepfakes” of photos and videos are much more sophisticated and designed to deceive. In the near future, we may not be able to trust what we see or hear. This leads to a very interesting rabbit hole of AI, verification.
Currently, our verification methods (DocuSign, two-factor logins, PIN codes) are designed to identify and verify senders and receivers of documents or data. With AI, there is the added step of verifying the authenticity of the data itself. This can be accomplished using blockchain technology, which has also been used to launch thousands of crypto currencies and NFTs (non-fungible tokens). NFTs function as an electronic “receipt” verifying the source of data. Used directedly, blockchain is not simply a way of verifying transactions or proving that you own an animated image of an ape, but a way to authenticate any form of data that you receive.
Although the field of artificial intelligence has been around for years, the rate of change in innovation and application has increased exponentially over the past 6 months and will continue to do so. AI is a subject that is changing so rapidly that much of what we’ve written might be moot within the next year. However, whether we recognize it around us or not, AI is going to become a much larger part of our everyday life than we could have imagined even two years ago.
Although investors are not yet able to directly invest in most pure play companies, like Open AI, we can participate through larger firms who can better deploy these emerging technologies. This is also a time to be wary of anyone tagging “AI” onto their name or mission statement – there are several parallels to be drawn to the similar trend with “crypto” or “blockchain” a few years ago. As always, the best way to play a new technology, or theme, in the market is to be widely diversified in that industry. We feel that we have well-diversified exposure to many components of the AI space through investments in software, hardware, and platform providers, and we look forward to future developments.
Shannon DermodyTEST