Thank you to Kem-Laurin Lubin, for sharing her insight in our knowledge base.

Click here to read on Medium.

Who really owns AI power and why you should care

“We need diversity of thought in the world to face the new challenges” — Tim Berners-Lee, the inventor of the World Wide Web

This week, TechCrunch reported that Emad Mostaque, the founder and CEO of Stability AI, has left to pursue interests in decentralized AI. The company now operates under interim co-CEOs Shan Shan Wong and Christian Laporte. To the average consumer of digital technology — no! let me rephrase this: it matters and let me explain why.

This recent development has prompted me to reflect on a topic that we seldom discuss, yet is crucial to our understanding of our position in the technological AI gold rush. It highlights our role as the foundational raw materials in this rapidly evolving AI rush. While I have touched on this subject previously, it’s imperative that we all acknowledge ONE critical, often overlooked reality of the AI revolution.

Our collective and cumulative human data serves as the essential raw material. We are the product, humans!

Earlier this month, I touched on the exponential growth of AI in a blog, highlighting its pervasive nature and how it has become so integral to our daily lives, it's almost like a utility. Specifically I quote myself from that post:

“Although not explicitly stated, at the heart of this blog’s inquiry lies the interplay between technology and its expanding array of seemingly indispensable applications, akin to civil utilities, which are gradually supplanting conventional lifestyles. One notable example is observed within the healthcare sector — there are now apps for that.”

Despite the considerable hype surrounding AI, it’s important to remember that its functionality hinges on aggregating vast amounts of data over time, harnessing our collective human experiences to produce results that significantly impact our lives. Yet, despite its widespread influence, the control over AI remains concentrated in the hands of a few powerful entities. This situation raises critical questions about the future direction of AI and its accessibility to the broader population as well as the use of us as product in this dialogic. And now as we witness a split in perspective by key players like Stability AI, I think it is time to create a conversation for us more regular folks.

Can AI and the data that informs it be owned?

And by whom? And pressingly for the purpose of this post, I want to discuss what the real story is here and why it matters.

The conversation to be had here is simple —should the path of AI development be centralized in the hands of a few or should it be decentralized where everyone feels ownership of their own AI-powered data?

Last year, I discussed how companies have frequently misused our data, often without our consent and straying from the original intent with which our data was shared. This is at the crux of my discussion today.

Centralized vs, decentralized AI — what is it?

Decentralized AI (DAI), as discussed in tech, represents a visible shift from traditional, centralized AI systems. Unlike centralized AI, where a few big organizations hold most of the power, DAI spreads this out to prevent problems like widespread surveillance, data leaks, and a few companies controlling everything. By using multiple locations for data and processing, DAI is safer and avoids having a single point that could fail.

External

DAI uses technology like blockchain to keep data safe and trustworthy, leading to a system where no single group can dominate. This setup increases privacy and fairness by using varied data and contributors. Also, decentralized AI can grow more easily by adding more points to its network, allowing for faster processing and quicker decisions without the slowdowns seen in centralized systems.

I explained blockchain here about 7 years ago — the link is here

Centralized AI (CAI) is where a single or few entities, like tech giants Google or Amazon, control or collective data and AI operations, processing all information on central servers to make decisions. Decentralized AI, on the other hand, distributes data and processing across a network, often using blockchain, like in a healthcare AI model where patient data is shared across many nodes. This means decisions in decentralized AI come from network consensus, not one central authority, enhancing privacy and security while giving individuals more control over their data.

Takeaway — Decentralized AI’s approach ensures that the system is more resilient to attacks and less prone to biases or manipulation by any single entity, promoting a more democratic and equitable use of AI technology.

Ok so what does that mean, applied?

Data portability the what and why

Inmy professional life, I operate within the highly regulated sectors of insurance and healthcare data in Canada, where safeguarding personal data is a paramount concern. In such environments, as customer apprehension regarding privacy escalates, questions arise about the implications of numerous companies forming partnerships and integrating their data systems. The critical issues of consent and purpose control come to the forefront. As a Design Strategist, I look into these matters from a risk and compliance viewpoint, examining how we can protect individual privacy while considering the complexities of data interoperability and corporate collaboration.

For instance, I scrutinize seemingly benign elements like consent workflows, questioning,

“Does the user understand, when they click that “Consent button?”

Do they understand explicitly what they are agreeing to, and how can they withdraw their consent?

Where are the UI affordances for the use to control their options?”

A user interacting with an interface might agree to share their data with a partner service, such as an online therapist. The question then becomes, how can they withdraw this consent after the session ends? As companies increasingly share data through partnerships to enhance user experience, it’s vital to keep the user’s perspective in focus, ensuring clarity on how their data is used and by whom.

Many companies still take the stance that it is their data to use as they see fit to grow their business. But is it the case?

So, here, in this instance data portability factors in and refers to the ability of that to easily transfer their personal data between different service providers, platforms, or applications.

Companies playing in this space — the future is decentralization

Data portability has become a key concept in privacy and data protection laws, such as the General Data Protection Regulation (GDPR) in the European Union, which gives European citizens the right to receive their personal data from one data controller and transmit it to another — frankly putting them in control. The technology is there are already possible and some of the companies in the space of data portability include:

Big Tech companies, including Google, Facebook (now Meta), and Apple, have recognized the importance of data portability and control, leading to the development of tools that empower users to manage their own information.

Google’s Takeout service, for instance, allows users to export a copy of their data from across Google’s products, such as emails, photos, and documents, enabling them to have a comprehensive view of the information stored about them by Google.

Similarly, Apple’s Data and Privacy portal gives users the ability to download a copy of their data associated with their Apple ID, which can include app usage history, purchase records, and more.

Facebook (Meta) has also introduced features that allow users to download their information, including uploaded photos, videos, and posts, providing a way to transfer data to other services if desired. These initiatives by Big Tech companies represent a shift towards more transparent and user-controlled data management practices, allowing individuals to take ownership of their digital footprints and decide how and where their data should be used.

Data management and integration companies such as Informatica, MuleSoft (a Salesforce entity), and IBM provide specialized services that enhance data portability and integration.

Informatica offers a suite of tools designed for comprehensive data integration, quality, and management, enabling organizations to aggregate, cleanse, and distribute data across various systems efficiently.

MuleSoft, under Salesforce’s umbrella, focuses on connecting different applications, data sources, and devices through its integration platform to facilitate seamless data flow and accessibility across enterprise ecosystems.

IBM also offers robust data integration and management solutions, leveraging advanced analytics and artificial intelligence to help businesses synchronize, replicate, and organize data across diverse environments. These firms play a pivotal role in enabling organizations to manage their data more effectively, ensuring it can be moved, shared, and utilized across different platforms and services to meet evolving business and compliance requirements.

Startups and specialized services such as Digi.me, Mine, and Solid, initiated by Sir Tim Berners-Lee, are at the forefront of empowering users to control and transfer their personal data across different platforms.

Digi.me allows individuals to consolidate their online data from various sources into a single location, providing greater control and insight into their personal information.

Mine offers a service that helps users discover and manage their data footprint across the web, enabling them to make informed decisions about where their information is stored and how it is used.

Solid, again, envisioned by Tim Berners-Lee, aims to revolutionize data ownership by enabling users to store their data in personal online data stores, or “pods,” that can be accessed and shared at their discretion. These initiatives reflect a growing movement towards enhancing data autonomy, allowing individuals to manage their digital identities and exercise control over their personal information in the digital space.

Taken together, all these companies provide various tools and services that enable individuals to manage, access, and transfer their data across different systems and services, promoting greater control and flexibility in the use of their personal information.

Like many of my colleagues, given our jobs in related spaces, we foresee this as the future trajectory for digital transactions, shaping how we interact in the online and digital world. But where does the end user come in?

Power to the people — putting you in control

Imagine your digital identity is made up of various data types, from healthcare records to social media profiles. Currently, this data is often controlled by the platforms that collect it, not by you.

Now, consider a different scenario where you hold the reins of your digital data. You decide who can access it and how it’s used. This change would drastically alter the landscape of digital ownership and privacy, offering a solution to the numerous data privacy issues we encounter today.

This concept is at the heart of the debate between centralized and decentralized AI.

Centralized AI systems rely on data collected and controlled by specific entities, whereas decentralized AI promotes a model where data control is distributed among users. This fundamental difference in data governance could significantly impact how AI systems are developed and used in the future. It also enables a more open, transparent, and competitive ecosystem for AI services. And, it would be a perfect use case for a decentralize model of use moving forward, giving power back to the people.

What are your thoughts about this topic?

References

1. Built In. (n.d.). The Case for Decentralizing AI. Retrieved from https://builtin.com

2. Gemini. (n.d.). The Convergence of AI and Crypto: Centralized Power Versus Decentralized Potential. Retrieved from https://www.gemini.com

3. Humans.ai. (n.d.). Decentralized AI, the new order in artificial intelligence. Retrieved from https://blog.humans.ai

4. Openfabric AI. (n.d.). Why Decentralized AI is better than Centralized AI. Retrieved from https://openfabric.ai

5. Techopedia. (n.d.). Decentralized Artificial Intelligence (DAI) | Definition, Features & More. Retrieved from https://www.techopedia.com

About me: Hello, my name is Kem-Laurin, and I am one half of the co-founding team of Human Tech Futures. At Human Tech Futures, we’re passionate about helping our clients navigate the future with confidence! Innovation and transformation are at the core of what we do, and we believe in taking a human-focused approach every step of the way.

We understand that the future can be uncertain and challenging, which is why we offer a range of engagement packages tailored to meet the unique needs of both individuals and organizations. Whether you’re an individual looking to embrace change, a business seeking to stay ahead of the curve, or an organization eager to shape a better future, we’ve got you covered.

Connect with us at https://www.humantechfutures.ca/contact