Did someone forward this to you?
Click here to join 31,879 receiving weekly tips via email and social.
Human Intelligence and Artificial Stupidity… and Vice Versa
Read time: 4.75 minutes
Welcome to Future-state Thinking, my weekly newsletter where I give actionable content, insights and tools for business and personal growth from my experience as an innovator and entrepreneur.
If you’re looking for my Cheat Sheets and Infographic PDFs, the vault is at the bottom of this email!
First, a word from my design thinking friends, Lucy and Tracy.
Their podcast is your key to unlocking creative potential and business growth!
Join Lucy & Tracy as they take you on your own hero’s journey – as you adventure through proven design thinking techniques to elevate your business and empower your life!
Subscribe to A Beginner’s Guide to Design Thinking today!
Hi Reader,
AI continues to shape how we think about intelligence, creativity, and problem-solving.
However, AI’s rapid evolution brings challenges, especially when we contrast its strengths with human capabilities.
This issue is based on a recent ‘raising the bar’ public talk from my friend and The DUCTRI Playbook co-author, Dr. Christian Walsh.
Today, we explore the tension between human intelligence and artificial “stupidity,” highlighting where AI falls short, how biases affect outcomes and humans’ role in optimising AI’s potential.
Let’s dive in!
The Intelligence-Stupidity Axis
To understand the interplay between AI and human intelligence, consider a framework that contrasts Artificial Intelligence (AI) with Artificial Stupidity on one axis, and Human Intelligence with Human Stupidity on the other.
This grid helps us map out where AI excels, where it struggles, and how human abilities intersect with AI’s potential.
This ties into Amara’s Law, which reminds us that we tend to overestimate the impact of technology in the short term but underestimate it in the long term.
While AI may seem overhyped today, its future impact is likely to be more profound than we anticipate.
As we explore AI’s strengths and limitations, it’s essential to remember that the long-term effects of AI may outstrip its initial breakthroughs. In other words, we think the development is linear, but it’s not.
Quadrant A: AI for Automation, Assistance, and Access
In this quadrant, AI addresses areas where humans typically struggle – tasks prone to errors, fatigue, or repetition. AI automates processes, making them more efficient and reducing human errors.
Automation and Assistance: A study conducted in a call centre using a GPT-based model demonstrated a 34% improvement in performance for new hires and underperformers, as the AI-assisted in routine processes.
However, top performers saw no improvement, with some showing slight declines in performance.
Additionally, a follow-up study revealed that employees trained to use AI (through prompt engineering) performed 27% worse in some tasks, highlighting AI’s limits in augmenting human expertise without clear understanding or oversight.
Questioning Efficiency: While AI can make processes faster, we should ask whether efficiency is always the right goal.
For example, improving a product to eliminate the need for frequent customer service calls may provide more long-term value than simply increasing the speed at which calls are answered.
This brings us to double-loop learning – a concept where instead of optimising processes (single-loop learning), we question whether the processes are necessary or valuable (doing the right thing rather than just doing things right).
Access: AI tools like visual design platforms and chat interfaces have dramatically lowered the barriers to entry, making advanced technologies more accessible to a broader audience.
This democratisation of AI allows more people to engage with technology creatively.
However, with this accessibility comes the need for caution in how AI is applied to ensure it’s being used to add real value, not just drive efficiency.
Quadrant B: Biases, Bot Shit, and Bad Actors
This quadrant highlights the risks of combining human and artificial biases. The intersection of these biases can amplify errors and expose vulnerabilities in AI systems.
Biases: Cognitive biases such as availability bias (where we overvalue information that is easily accessible) and confirmation bias (where we seek out information that aligns with our existing beliefs) shape human decision-making.
In Thinking, Fast and Slow, Daniel Kahneman explains how these biases stem from our reliance on System 1 thinking – the brain’s fast, intuitive process. While heuristics are valuable, AI inherits and amplifies these biases.
Further complicating matters, biases like fluency bias (the tendency to believe well-articulated ideas) and the illusory truth effect (where repetition makes a statement feel true) make AI-generated content even more convincing, despite its potential inaccuracy.
Bot Shit: A significant issue with AI is its ability to produce convincing but false information.
Large language models such as GPT often generate content designed to engage users, but not necessarily grounded in truth.
This “bot shit” phenomenon presents a challenge: AI fluency can mask underlying inaccuracies, making the information sound credible even when it’s not.
AI-generated outputs are designed to keep users engaged, but the potential for misinformation is high without human oversight.
Bad Actors: The misuse of AI for malicious purposes is another growing concern.
In 2022, Meta reported shutting down 5.8 billion fake Facebook accounts created to spread misinformation and manipulate opinions.
While this number dropped to 3 billion in 2023, the sheer scale remains alarming.
AI-driven bots, spam accounts, and fake profiles are tools frequently used by bad actors to influence social platforms, highlighting the need for continuous vigilance.
Quadrant C: Curiosity, Creativity, and Connection
This quadrant focuses on the human superpowers – areas where AI struggles. Curiosity, creativity, and connection are critical elements of human intelligence that AI cannot fully replicate.
Curiosity: One of the biggest differences between humans and AI is curiosity.
AI can respond to prompts and follow instructions but doesn’t ask why or challenge assumptions. Human curiosity leads us to dig deeper, uncover root causes, and explore new possibilities.
This is particularly evident in problem-solving and innovation, where human inquiry drives deeper understanding.
AI lacks this instinct, limiting its ability to navigate complex or nuanced problems.
Creativity and the Creative Cliff Fallacy: Creativity is the ability to generate ideas that are both novel and valuable. While AI can produce novelty, it struggles with understanding what is truly valuable in a human context.
A study by Boston Consulting Group (BCG) examined how consultants used AI for two distinct tasks: creative problem-solving and business problem-solving. For the creative problem-solving task, AI use led to a 40% improvement in performance.
However, the study found that AI tended to generate many similar ideas, lacking the diversity and out-of-the-box thinking that humans naturally bring to creative tasks.
Conversely, for the business problem-solving task, AI actually reduced performance by 23%, particularly in tasks that required interpreting qualitative data and human nuance.
Another study from Stanford found that while AI helped generate a high volume of “B” and “C” grade ideas, human teams produced more “A” grade (high-value) and even “D” grade (low-quality) ideas, reflecting a wider diversity of thought.
This highlights the creative cliff fallacy—the idea that our best ideas come early in the creative process. In reality, research shows that deeper, more valuable ideas often emerge later, after sustained engagement and incubation.
Diversity of thought is critical for creativity and where humans excel.
AI tends to generate ideas quickly, but human creativity thrives on time and persistence, producing higher-quality results the longer we stick with creative tasks.
Connection: Connection, particularly emotional connection, is another area where humans excel.
AI can process and analyse vast amounts of data but cannot form the meaningful human connections necessary to drive empathy, understanding, and relationship-building.
These connections are crucial in contexts like leadership, customer relations, and teamwork, where understanding people’s needs and emotions is critical to success.
AI, while a powerful tool, lacks this relational capability.
Quadrant D: Optimising for the Future
Looking forward, AI has the potential to complement human creativity and intelligence, but it must be used thoughtfully.
The future lies in optimising AI’s strengths while leveraging human capabilities to fill in the gaps.
Optimising Creativity: AI can assist with tasks that require speed and repetition, but humans must guide the creative process to ensure diversity and innovation.
The long-term value of creative problem-solving often comes from sustained effort.
As the creative cliff fallacy suggests, sticking with creative tasks over time leads to more innovative and valuable outcomes.
This also ties into double-loop learning – humans must question the underlying assumptions about why specific tasks are done, rather than simply optimising how to do them.
An Optimistic Future: While legitimate concerns about AI’s limitations and potential misuse exist, there is a clear path forward.
AI can assist us in areas where we are weak, but human creativity, judgment, and emotional intelligence remain essential to solving complex problems.
The future of AI is about collaboration – using AI to augment human strengths, while remaining critical of its weaknesses and careful about its applications.
Created using ChatGPT
⚡️ Call to Action
AI’s potential lies in its ability to complement human intelligence, but it’s critical to understand where its strengths and weaknesses align with our goals.
As Amara’s Law reminds us, the long-term impact of AI may be far more profound than its initial breakthroughs.
The future of AI should focus on optimising collaboration between humans and machines, ensuring that AI enhances creativity, innovation, and productivity without compromising the value of human insight.
Key Actions:
- Use AI for Repetitive Tasks: Automate where possible to free up time for more creative and strategic work.
- Enhance, Don’t Replace: Use AI to support decision-making but not rely on it to replace human judgment in critical areas.
- Challenge Biases: Be aware of how AI can reinforce biases, and work to challenge and correct these where possible.
- Stay Curious: Leverage human curiosity and creativity to dig deeper into problems, and use AI as a support tool to broaden the scope of your work.
Are you optimistic about the future of AI?
Vaughan’s Vault:
P.S… As promised on LinkedIn, click the button for my cheat sheets on innovation, strategy, and more!
Did someone forward this email to you? Sign up here.
P.S. Want to unlock your innovation and impact potential even more?
There are a few ways we can help:
- Grab your super handy resource. Join the waitlist for our new DUCTRI Playbook for a new way to inspire curiosity, unleash creativity and ensure clarity when implementing solutions.
- Adapt to hyper-disruption. Join the waitlist for the new DUCTRI Mater Course. Where driven professionals become skilled human-centred innovators and change-makers.
- Need help now? Contact me here for design thinking consulting, workshops and training (limited availability).
Pssst. Lucy gave last weeks issue a 5-star review with this testimonial:
“I learned a new term today – process debt. Ironic as that’s exactly what I’m knee deep in with a client project atm. Another great newsletter Vaughan! “🙏
Thank-you Lucy!
Share With Friends, Get Rewards!
Do you know someone who would love this newsletter?
Invite them to signup using your trackable link below. Or share directly on social or by email.
Twitter
Linkedin
Email
See what you could win >
⚡️ by SparkLoop
