- Work3 - The Future of Work
- Posts
- Let AI help you...think
Let AI help you...think
How AI can improve our thinking and reasoning
This week, I have a hot take 🔥.
I think one of the best use cases for GenAI is not to take away work, but to help humans think better. This would mean we can use the technology to:
Act as a soundboard to test our ideas
Find counter-arguments to our points of views
Practice iteration, problem-solving and the Socratic method of learning
Let’s dive in! 🤿
What we thought AI was going to do
Whilst at the beginning of the hype cycle (here below you can find Gartner’s) for AI we had inflated expectations, and still do, about completely removing some jobs, as time passes by we’re coming to terms with its (current, but likely permanent) limitations.
However, I’m starting to think that we’re looking at the finger, rather than the moon here. Take the fundamental examples of GenAI being used to write code and copy.
Yes, you can get some good output. It could be even great, for basic needs and types of tasks. I’m not arguing that some tasks will go away. It’s happened many times before (our human brain quickly forgets…just look at how fast it’s happened with the pandemic).
But let’s apply my beloved ‘Second-order thinking’
Second order thinking is the ability to consider the potential consequences and implications of a decision or action beyond the immediate and obvious outcomes. It involves thinking beyond the first level of thinking and considering the second and third order effects. This type of thinking requires a deeper analysis and understanding of the interconnectedness of various factors and the potential ripple effects that may occur. By engaging in second order thinking, individuals can make more informed and strategic decisions, anticipate potential risks and benefits, and avoid unintended consequences.
GenAI will have mass adoption → Basic Content / Code creation will be cheap → Higher demand for complex and human thinking
Now, stay with me. I’m looking for a way to explain this, I’m here on my laptop writing this on my own. Who do I turn to?
Kevin Roose (columnist at The New York Times, and Podcast host of Hard Fork - great one to check out if you haven’t) said ChatGPT’s #1 use case is that it’s the best research assistant he’s ever had.
That’s 100% true, though of course there is a fact-checking issue to be done (incorrect statements on GenAI happen between 4.8% and 26% of the time, depending on the statement category), and be careful to watch out for next week’s launch of OpenAI’s Google competitor.
My argument is that its #1 case, which can be a fundamental enabler of the future of work, is it can help you to think.
Take my example. I’ve started this newsletter for two reasons. 1) I wanted to build a product for ‘future of work', but didn’t find the right idea. Frustrated, I thought: let me write about it 😂. 2) which I realized only in hindsight:
I needed to improve my thinking skills. A lot.
Why writing acts as a powerful tool for thought in several ways
Externalizes thoughts: Our thoughts can be fleeting and jumbled. Writing them down puts them in a tangible form, allowing us to see them more clearly and objectively. This externalization helps us analyze, refine, and develop our ideas.
Forces organization: Writing requires us to structure our thoughts logically. We need to consider what comes first, what evidence supports our claims, and how to connect ideas smoothly. This process strengthens critical thinking skills and clarifies our understanding.
Deepens engagement: Writing makes us delve deeper into a topic. As we elaborate and explain, we identify gaps in our knowledge and areas that need further exploration. This active engagement leads to a more comprehensive understanding.
Promotes discovery: The act of writing itself can be surprising. Sometimes, as we express our ideas, new connections and insights emerge. Writing can help us discover new aspects of a subject, or even ourselves, that we weren't aware of before.
Enhances memory: Writing down information helps solidify it in our minds. The physical act of writing, particularly by hand, strengthens memory pathways more than simply typing This improved memory allows us to access and utilize information more effectively.
Here’s the interesting part: In essence, writing is a dialogue between ourselves and the page. It's a process of exploration, clarification, and discovery that leads to a deeper understanding of ourselves and the world around us.
A dialogue. Rings any bells?
Why Dialogue is the “Building Block of Thought”
Social beings by nature: We are inherently social creatures and evolved to rely on each other for survival and thrive through communication and collaboration. Dialogue isn't just a way to share information; it's how we test ideas, refine thoughts, and build understanding.
Thinking through talking: When we talk, we externalize our thoughts. This process forces us to organize our ideas, identify logical gaps, and refine our arguments. The back-and-forth of dialogue allows us to see our thoughts from different angles, leading to a deeper understanding.
Learning from others: Through dialogue, we learn from others' perspectives, experiences, and knowledge. This exposure challenges our assumptions, broadens our understanding of the world, and sparks new ideas. Imagine scientific breakthroughs that emerged from debates and shared observations.
Thinking is what has enabled us as a species to adapt, survive, and thrive. It’s what has led to AI and what will lead to whatever’s next.
Yet, who teaches us how to think?
Our parents, schools, sure. But ultimately, we’re all different.
According to the Flynn effect, each generation gets smarter
According to the Flynn Effect, each generation is getting smarter. But IQ tests focus on specific abilities, not overall thinking capabilities. So someone with a high IQ might struggle with real-world decision-making.
IQ tests measure things like logic, abstract reasoning, and working memory. These are important cognitive skills, but they don't capture the full picture of intelligence.
Real-world decision-making requires additional skills like:
Critical thinking: Evaluating information, identifying biases, and considering different perspectives.
Emotional intelligence: Understanding your own emotions and those of others, using emotions to guide behavior.
Problem-solving: Identifying problems, generating solutions, and evaluating their effectiveness.
We're living in a world of information overload. Thanks to the internet and AI, finding facts and figures is easier than ever. However, this doesn't necessarily mean we're becoming better thinkers.
There's a growing concern that critical thinking skills, like analyzing information, identifying bias, and forming strong arguments, might be on the decline. This is a crucial skillset for the future of work, because AI can't replace our ability to think creatively, solve complex problems, and make sound judgments.
The good news is, AI can be a powerful tool to assist us in the thinking process:
AI is built around simulating human conversation. Chatbots, virtual assistants, and language models all thrive on their ability to engage in a back-and-forth exchange, similar to how humans learn and think.
My go-to is the Socratic Method: a cornerstone of critical thinking, that involves asking probing questions to stimulate self-discovery and expose hidden assumptions. AI can supercharge this method by assisting in crafting powerful questions, acting as a real-time research assistant during discussions, and even facilitating debates from opposing viewpoints. This AI-powered Socratic dialogue can lead to a more thorough exploration of ideas and a deeper understanding of complex issues.
Here's an example of how this might work in practice:
Imagine you're considering a career change. You can:
Use an AI idea generation tool to create a list of Socratic questions about your current job satisfaction, skillset, and desired career path.
As you ponder these questions, utilize AI to research potential career options that align with your answers.
If you're stuck, engage an AI debate platform to present arguments for and against different career paths, prompting further self-reflection.
Finally, analyze the conversation with AI to identify areas where your knowledge might be lacking (e.g., specific skills required for a new career) and use this insight to guide your next research steps.
Last, but not least, a word on ‘iteration’.
The real power of AI and ChatGPT is that it can quickly give you new forms of output. So in this dialogue, you’re constantly tweaking and optimizing to obtain your desired outcome. This can be frustrating, and at times not even yield what you want. I think we need to start to see this as a new way to think and learn and see the opportunities and risks that it poses since it’s likely not going away anytime soon.
In 2013, the Project Management Institute concluded out of their data that one out of five projects fail because of poor communication within a team environment. But don’t think you’re safe when you’re working alone.
Ask yourself: how many times have you blocked yourself because you did not document/explain the way you made something?
Here’s three practical ideas I’ve been using:
Challenge your assumptions:
Ask an AI to generate arguments opposing your current viewpoint on a topic. This can expose blind spots and encourage you to consider alternative perspectives.
Explore possibilities:
Describe a problem you're facing with an AI and have it brainstorm potential solutions. This can open your mind to new approaches and creative ideas.
Practice complex reasoning:
Engage in debate-style simulations with AI. By responding to its arguments and counter-arguments, you can hone your reasoning skills and build a stronger case for your ideas.