“Whoever wins the personal agent, that’s [going to be] the big thing.  Because you will never go to a search site again, you’ll never go to Amazon again” - Bill Gates ​

By dr. Paul Jurcys

In the very near future, each of us will our personal AI assistants for every aspect of our lives. Each of us will have our personal AI doctor, personal AI coach, personal AI nutritionist, or digital assistant that will help us organize our calendars, schedule events, and purchase items on our behalf. 

Such AI-powered assistants will augment our abilities and help get rid of time-consuming tasks, and focus on things that matter most. 

• What are the current challenges to make such digital assistants really personal? 
• What technology infrastructure should be in place to make our personal AI assistants really personal? 
• How will these assistants interact with each other? 
  ​​

In this post, we explore (i) the current trends in the market, (ii) the types of personal AI assistants and (iii) emphasize the need to adopt a human-centric approach to data in order to build truly personal AI that runs privately, on top of your own data.

Buzz In the Market

In the past few months, there has been quite a bit of excitement about the possibility of building truly personal AI: 

• InflectionAI raised $1.3 Billion to build personal AI assistant “Pi”
• Jason Calacanis sees Personal AI as a $20 trillion opportunity
• Doug Clinton Suggests that truly Personal AI should not be built on “centralized” models
• Doug Clinton also conducted a poll and found that individuals are willing to pay for their Personal AI. 
  
  

Investors start to realize that Personal AI is a new market that both legacy players such as Microsoft and Apple, as well as startups are trying to explore. 

A massive investment in InflectionAI could be seen as an attempt to stave off any potential competition and secure dominance in this rapidly emerging domain. 

Shortage of Experts
Today, we are unfortunately facing a remarkable shortage of experts. With over 8 billion people on Earth, there are a mere 10 million doctors. Teachers struggle to spend enough time on direct interactions with every student in the class. 

On a personal level, we often lack someone equally knowledgeable or passionate about our favorite topics. In times of grief, suffering, and emotional pain, the desire for a close companion to provide advice and understanding becomes ever more acute.

This is where the transformative power of AI comes in. In the near future, countless individuals, potentially everyone, will be able to augment their personal capabilities with our own personal AI assistants. 

The trajectory toward universal access to intelligence is clearly mapped out: we can expect that any person with access to the internet and a hand-held device will have access to the same expert advice from a doctor, an educator, or any other specialist, all through their personal AI. 

2 Ways to Build Personal AI
As technology continues to leap forward at ever-increasing speed, personal AI assistants are evolving to become more sophisticated and versatile. 

There are two paths of bringing AI superpowers to humans: 

A top-down approach: after the release of ChatGPT and other generative AI tools, we are witnessing an influx of various personal AI assistants that are supposed to hello us find information, summarize text, or generate images. 

The top-down approach is rooted in creating software applications - personal AI assistants - that are marketed as tools to help us solve various daily tasks and assignments or offer us a companion, a buddy that is available 24/7. Think of InflectionAI’s Pi, Midjourney, Dalle, and countless others. 

However, most of our essential tasks are rooted in the physical world: we need to eat, sleep, and and go to the bathroom. As cheap sensors are being integrated in our wearable devices and environments around us, individuals have tools and ability to track personal biological clock - we start living quantified lives. If you are a runner, you probably have a fitness tracker to measure your running distance and heart rate, as well as recovery rate. Also, our cars track hundreds of data points (e.g., acceleration, braking, engine performance, and outside factors). 

Scientists around the world are building assistive robots to help us to recover from injuries or learn important skills.

These sensor-derived data sets also lead to the development of AI-powered applications that help optimize different aspects of our lives. This sensor-data-AI path resembles a bottom-up approach to building personal AI. 

Data Eats the World, But How to Access Data? 
AI assistants such as ChatGPT, or PI, are amazing in answering our general questions and performing general tasks. However, if you ask anything personal, something related to yourself, they stumble. Let’s see what responses we get to the question “How did I sleep last night?”:

How come!? Isn’t it odd that if you wear Apple watches that track our activity, sleep and process various biometric data, Siri - an AI assistant built by Apple for Apple’s products - is not able to answer such a simple question?!

Eventually, every company building personal AI will face the challenge of connecting the personal AI to the user’s own data.  
​
To build a truly personal AI, we need not only general knowledge from publicly available sources and research repositories, we also need to bring this general, publicly available knowledge and correlate it with the specific data of each individual. In other words, to build a truly personal AI, Ai assistant must be able to access to the user’s personal data. Otherwise, the AI assistant will be a generalist, not a personal: ​

On July 13, 2023, an event called “Augmenting Consumer Experiences in the Age of Data & AI” took place at the offices of New York Life Insurance Company's offices in San Francisco (kindly hosted by Boudjemaa Nait Kaci).

​Augmented UX with AI

The event started with Jouko Ahvenainen from Prifina, who invited his fellow panelists to delve into their personal experiences, both triumphs and trials, relating to their interaction with the recently released AI and data-driven technologies.

Heather Whitney from Morrison Foerster shared her excitement at the launch of ChatGPT. She underscored the versatile potential of generative AI, highlighting its role in revolutionizing customer experiences across a variety of sectors, notably content creation, gaming, as well as gene editing and drug development. 

Following up, AJ Tomas from Touchdown Ventures shed light on the latest unveiling from Humane Inc. — a secretive startup founded by former Apple executives — who have recently announced the forthcoming release of a cutting-edge gadget that generated quite a buzz in the tech world.

Amit Sharma shared his vision: 

"We believe that generative AI has the potential to create innovative forms of content that go beyond the current video-based experiences on Instagram, YouTube, or Tiktok. These novel experiences will offer users the opportunity to engage with brands and services in a multimodal manner."

Aaron Mollin (CEO, Ichijiku) offered a fascinating observation, arguing that most AI-fuelled experiences are curated from a “top-down” perspective. He recognized the transformative nature of tools such as ChatGPT, but called to realize that oftentimes he felt that such tools are detached from how we interact with material objects around us. 

Aaron shared an alternate viewpoint related to integrating AI-powered insights based on the data collected from ubiquitous sensor technology found in everyday physical objects, like wearable tech, IoT devices, and environmental sensors. He called this the "bottom-up" approach. One such example is a collaboration between his company, Ichijiku, and Prifina: their work has resulted in the creation of sensor-imbued luxury jackets (one of which he was wearing that night).

This conversation immediately led to a conversation with audience members about the fact that we, individual consumers, all have cell phones, and we find it cumbersome to interact with hundreds of separate apps and services. From the consumer perspective, rather than multiplying points of interaction, it would be better to reduce our reliance of more and more devices and apps and make the UX more user-centric. 

Predicting the Future 

Jouko then asked panelists to share their thoughts on what changes consumers could expect in the next coming years.  

Amit did not hesitate to suggest that every aspect of consumer experience will change in the next few years. Some UX aspects will change faster than others, but gradually many industries will transform. E-commerce, online shopping, entertainment, and interactive experiences are the first ones to transform. 

AJ Tomas predicted that another area of fast-moving development is personal productivity. 
Aaron expressed his wonder about the impact of data on the way how individuals make decisions about their own lives. Lots of work appears to be needed to build personal recommendations and nudges for people. 

Jouko reiterated his favorite idea about predictions: although we do not have a crystal ball, we can actually make quite certain predictions about the future. Jouko noted that we can anticipate that certain things will happen (e.g., self-driving cars), but the challenge is to identify the timing: will happen in a year? Two years? Five years? 

Amit noted that the current developments in generative AI capture much attention of billions of people who are waiting to see what the actual breakthrough will actually look like:

“And if you go back to the 1980s, people knew there would be something called ‘a computer.’ But nobody knew what a computer would look like … it wasn't clear that it would be a box on your table with something called a keyboard and a mouse… This is true for smartphones and, most recently, Apple VisionPro. We can’t even imagine what Apple VisionPro 14 will look like. And that uncertainty is what makes people excited.” ​

Legal Issues To Be Resolved

Heather shared insights based on her hands-on experience from working with generative AI technology companies; she provided an overview of some of the key legal issues that revolve around copyright, the legality of the use of data, and data privacy. These issues are currently unclear and are currently discussed among legal experts and AI entrepreneurs:

• The legality of data scraping and using such publicly available data to train machine learning models: Do you have to license that data? Or can you scrape the entire internet and use that? Is that going to be deemed as copyright infringement and thus not permissible? Could it be excused under the copyright law doctrine of fair use? 
  ​
• Data privacy: what if the data scraped from the internet contains some personal information of individuals? Could such data be used to train ML models? Which laws should be applied to such activities (the law of the state where the allegedly infringing  AI company is based or is operating? The law of the place where individuals whose personal information is use? Or some other country’s law?). 
  
  
• Then there's the question about the model itself. An understanding of what they receive from its training data. People don't fully understand that yet. And engineers don't understand exactly how to describe that yet. And what it really means from a legal perspective is still in flux. They just don't know. And it is not currently 100% knowable. 
  
  
• The legality of outputs generated with AI tools: what is the relationship between your training data and the outputs? Could output results be considered to be “derivative works” (in a copyright law perspective) of the training data? If so, then it is possible that there will be massive restrictions on how those outputs can be used. What about the commercialization of these models? The use of outputs are likely to be curtailed as long as there is no clear answer whether the outputs are not infringing (copyright law). 
  
  
• Rights to outputs generated with generative AI tools. Having (any) rights to the output generated with the help of AI tools is a huge problem that will determine whether such works can be commercialized. 

Heather and Jouko shared their views about the emotional narrative around these generative AI tools: many artists feel hurt and believe their works have been unlawfully used to train AI without their permission. 
​
The ethics of this debate raise controversial questions. On one hand, information in the public domain is supposed to be free. On the other hand, some stakeholders invoke strong arguments based on the ideas of exclusive ownership rights. 

Ichijiku: Combining Fashion, Cultural Expressions and AIAaron shared his unique perspective on using traditional kimono silk materials to create bespoke fashion items. 

Differently from the prevailing ideals in the luxury fashion industry, where consumers are made to feel that the luxury fashion items purchased just a few months ago are out of style and invited to purchase the next generation of items, Ichijiku aims to create life-long items that people could hold on to for their entire lives.

Ichijiku embeds sensors in those fashion items and uses data to extend the lifetime quality of the garments. By implementing sensors and generating data, Ichijiku jacket owners can monitor explore to humidity, sunlight, and heat; the first, humidity, being the most kind of serious culprit for the degradation of silk. 

By using data and AI, Ichijiku aims to create a new type of experience that go even beyond the maintenance of one-of-its-kind silk. Additional information collected from other sensors, such as heart rate, could help people get insights into how they feel while wearing Ichijiku’s bespoke jackets. This is something that no one has ever been able to create. 

“For me, it's just all about, creating pieces that will stand the test of time. … Ichijiku aims to create these unbelievably beautiful, arguably one of the most intrinsically valuable items on the planet. And that I think, is what's going to really appeal to people. Technology will never be the selling point, although it is definitely appealing. We're using vintage silk fabrics, which ties closely to sustainability ideas. People understand just how beautiful the items they are buying. The use of data and technology helps us further the story.” -- Aaron Mollin

Post-Event Networking 

One of the remarkable features of the data and AI-related events currently happening in San Francisco is that people come to meet one another, learn about each other and what others are building. 

The same spirit could be felt at our event: attendees spent the remainder of the evening at our venue socializing, getting to know each other, and connecting on various social media platforms to stay in touch in the future. 

We at Prifina are excited to be at the heart of this historical moment; we feel inspired and motivated to continue building our Human-centric data platform for AI-powered apps and services.