Intro
As our world becomes increasingly digitized and reliant on artificial intelligence (AI), I had a thought. It is crucial to consider the implications of AI access on society.
Picture this: you wake up one morning, grab your phone to check the news, and realize that you can’t access the internet. Frustrating, right? Now imagine a world where access to artificial intelligence (AI) is just as crucial as internet access, but not everyone who wants it, has it.
As AI becomes more integrated into our daily lives, we need to start thinking about whether access to AI should be considered a basic right, just like the internet has (mostly) become as it parallels the right to free speech or the right to assemble. Because let’s face it, in an AI-driven world, those without access to AI tools and technologies are going to be left behind. Am I pre-emptive on this? No, I don’t think so.
The Right to Knowledge and Reasoning
The right to internet access is often viewed as an extension of the right to knowledge. In a world where information is readily available online, those without internet access are at a significant disadvantage when it comes to educating themselves and making informed decisions. Similarly, as AI becomes more integrated into our daily lives, those without access to AI tools and technologies may find themselves at a disadvantage in terms of reasoning, problem-solving, and decision-making. I can attest that I am far more efficient when augmented and working in collaboration with AI as a value-multiplier.
And we’re not just talking about any old AI system here. We’re talking about the top-of-the-line, the frontier models. Because let’s be real, having access to a basic chatbot is not the same as having access to the most advanced AI models out there. It’s like showing up to a Formula One race with a tricycle. Where anyone with even basic internet access can load a Wikipedia page, albeit slower, the value is the same. This is not the case from model to model with AI.
Equal Opportunity and Competition
One parallel between the right to AI and other fundamental rights is the right to compete on an equal footing. Just as access to education and resources is essential for individuals to compete in the job market and contribute to society, access to AI tools and technologies may soon become a prerequisite for success in many fields. Without equal access to AI, we risk creating a new form of inequality, where those with AI access are able to outperform and outcompete those without.
This inequality could have far-reaching consequences, from limiting economic opportunities for individuals to stifling innovation and progress at a societal level. As AI becomes more integrated into various industries, from healthcare to finance to education, those without access to these tools may find themselves left behind, unable to compete or contribute to the same extent as their AI-augmented peers.
Impartial Mediation and Algorithmic Decision-Making
Another parallel between the right to AI and other fundamental rights is the right to impartial mediation. As AI systems become more advanced and are used to make decisions that affect our lives, from loan approvals to criminal sentencing, it is essential that everyone has access to these systems to ensure fairness and transparency. Without equal access to AI, we risk creating a system where decisions are made by algorithms that are biased or lack accountability.
To address this concern, organizations like Anthropic are working on mechanistic interpretability, which aims to decompose language models into understandable components. This approach could help ensure that AI systems are making decisions based on transparent and explainable reasoning, rather than operating as black boxes. By ensuring that everyone has access to AI tools and technologies, we can enable individuals to employ these interpretability techniques when a decision is made that affects their lives, keeping AI systems accountable and transparent. https://www.anthropic.com/news/decomposing-language-models-into-understandable-components
Cognitive Liberty and the Right to Augment
The right to AI can also be seen as an extension of the right to cognitive liberty. Just as we have the right to freedom of thought and expression, we should also have the right to augment our cognitive abilities through the use of AI tools and technologies. Denying individuals access to AI could be seen as a form of cognitive oppression, limiting their ability to expand their knowledge and reasoning capabilities.
As AI becomes more advanced and integrated into our lives, the ability to augment our cognitive abilities through AI will become increasingly important. Those without access to these tools may find themselves at a disadvantage not just in terms of competing in the job market, but in terms of participating fully in society and realizing their full potential as individuals.
I see denying this is similar in many ways to denying access to a calculator for math, or saying that Wikipedia is an unreliable source without taking into account the historical versioning and thus, semblance of peer review transparent in the articles, but alas, that might be a tangent and pet-peeve.
Okay, so what are some ideas for a solution?
To ensure that everyone has equal access to AI, we must work towards implementing solutions at both the technological and policy levels. One potential solution is to invest in infrastructure, similar to what Starlink aims to do for internet access, to ensure that AI technologies are available to all. This could involve developing more affordable and accessible AI tools and platforms, as well as investing in education and training to ensure that individuals have the skills and knowledge needed to use these tools effectively.
At the policy level, government regulation could play a role in ensuring equal access to AI. But let’s be real, government regulation isn’t always the most efficient solution, especially when it comes to keeping up with the lightning-fast advancements in the field of AI. As such, it may be necessary to explore alternative approaches, such as multi-stakeholder initiatives that bring together industry, academia, civil society, and government to develop standards and best practices for AI access and governance.
This is where open-source models come in. In combination with the aforementioned, I would personally weight this this highest for efficacy. By making frontier AI models open-source, like Grok being released and, more recently, Llama 3, we can ensure that everyone has access to the same advanced AI tools and technologies, regardless of their background or resources.
Conclusion
At the end of the day, the right to AI is about more than just access to technology. It’s about ensuring that everyone has the opportunity to participate in and benefit from the future of knowledge and reasoning. And the key to making that happen is through the development and promotion of open-source AI models that operate at the same level as the most advanced models out there.
By making these frontier models open-source, we can level the playing field and ensure that everyone has access to the same powerful tools and technologies. No more AI-haves and AI-have-nots, just a world where everyone can tap into the full potential of AI and use it to make their lives better.
So let’s get it done. Let’s work together to create a future where access to AI is a basic right, not a luxury. Let’s try to lessen the growing pains we experienced (and still are in many cases) with internet access. Where everyone has the tools they need to succeed and contribute to the greater good. And where the most advanced AI models are open-source and available to all, rather than being locked away in the ivory towers of Silicon Valley, et al.
Because in an AI-driven world, access to knowledge and reasoning shouldn’t be a privilege, it should be a right. And with the power of open-source AI models, we can make that right a reality for everyone.
If you want to work with me, feel free to reach out
ardea.io | john@ardea.io | https://www.linkedin.com/in/johnathangardner/ | https://github.com/Sinjhin
