Streamlined decisions | Systems from the AI dystopia series
Efficiency vs. freedom: can AI streamline our lives without sacrificing our choices?
Hello,
I am running a new experiment, “The Wild Intelligence Bench,” to build high-performance systems in the context of AI dystopia, and I’d like you to be involved.
In the upcoming editions of the AI dystopia series, I will assess featured systems across three primary dimensions": functionality, usability, and reliability.
I will later explore the second batch of dimensions: security, scalability, and cost.
Let’s focus first on why I do that. As mentioned many times in this publication, we humans can genuinely understand information and concepts, while LLMs are very good at recognizing patterns in massive amounts of text data. This lets us generate coherent text, but we don't grasp the meaning like humans do. AI can analyze vast amounts of data to identify correlations and trends that we might not see on our own—else way said link the unlinkable, make nonobvious appear obvious, or see the unseen.
Let’s consider the five stages in the evolution of artificial intelligence: science fiction, AI (some will understand it as “non-deterministic computing”), 'just software', digital transformation, and legacy. During every transition between these stages, confusion and errors reign. However, as for everything, we only test and learn during practice (implementation, integration, or development).
“Education is the most powerful weapon which you can use to change the world.”—N. Mandela
The problem—more often than not—is that people avoid accepting “evaluations” or “rates” as part of the transition processes. These are not information retrieval systems, not answer machines, and shouldn’t be considered linear thinking, expecting precise outputs like from databases.
Working with AI/ML requires a shift to a more flexible, probabilistic mindset, appreciating its creative and contextual capabilities despite occasional errors.
Here’s to help you learn about AI and its potential benefits and risks. You will receive an actionable system insight every week. So you can engineer the three main recommendations provided in the AI dystopia series:
Ask difficult questions.
Engage with decision-makers and hold them accountable for responsible AI development.
Look for organizations promoting responsible AI development and lend them your voice.
Unlike your typical review, I focus on factors that influence our decision in the AI era so you will get to know (and be involved) in the conception of our evaluation system.
About the AI dystopia series
Part 1: The genesis: a flawed utopia
Part 2: The algorithm’s bias
Part 3: The singularity and its shadows
Streamlined decisions
Imagine a world where every decision, big or small, is made for you. No more agonizing over career paths, fretting about commutes, or stressing about life choices.
But this utopia comes with a twist: an all-powerful AI governs every aspect of your life.
Intriguing, isn't it? Would such a streamlined existence be a dream come true or a descent into a chilling dystopia?