Communicating with Machines: A Personal look into Our AI Future

(Editor’s Note: If you are at all curious about AI, Artificial Intelligence, and don’t know where to start, here it is thanks to Marsh Creek resident Gerry Kurth who has done a magnificent deep-dive into the subject to get you on your way).

By Gerry Kurth

Like many of us, I’ve watched technology evolve over decades, watching computers move from rooms to pockets. But when I first encountered AI in early 2023, something felt fundamentally different. This wasn’t just another gadget upgrade. This felt like a technological discontinuity… maybe up there with the industrial revolution, electricity, or the internet. Could the talking heads be right?

To be clear, I don’t come with credentials in artificial intelligence or a PhD in machine learning. My background is engineering and business, but I quickly became an industry observer who got very curious. What started as a casual weekend comparison between AI and Google search turned into quite an exploration. With only 10-20% of us using AI regularly, I wanted to share what I discovered so far – because I have a feeling that we’re on this journey together.

We begin with hands-on “use cases” and the takaways of amazing capabilities and serious concerns. It’s about a 5 minute read. There is an add-on section that asks 6 fundamental questions about this technology that “could change everything.” Each question merits its own deep dive, just not today. For additional drill-down, I’m providing readers with a short list of references.

Beware that this piece reflects my enthusiasm for AI’s potential more than the critical analysis it deserves!    

The Great Face-Off: Google vs. AI

I began with a simple test: comparing Google search with ChatGPT using my typical half-formed, typo-ridden queries. Even with serious typos and illiterate questions, AI rephrased my intent clearly—”It sounds like you’re asking about…”—then answered what I meant, not just what I typed. The old Google corrected typos but often missed the heart of my question.

The real surprise came when ChatGPT asked me a clarifying question to better understand my intent. That was something Google had never done. This shifted the interaction from simple data retrieval to collaboration.

The contrast grew sharper with voice input. I could ramble half-formed thoughts, including pauses and “thinking voids.” The AI took my verbal mess, rephrased it coherently, and provided structured responses—something fundamentally different from any search engine I’d used.

How Have I Used AI So Far?

Medical Assistant: I fed my annual blood test results—a jumble of reports from different years and labs—to the AI. It aligned markers across time, identified trends, and compared them to age patterns over time. It flagged one marker which had drifted outside normal ranges. My doctor confirmed the finding on my next visit (though deemed it insignificant). It was like having a caring medical assistant to prepare for your 8-minute appointment.

Digital Paralegal: A 20-page legal document I was working on got the AI treatment. It flagged several internal inconsistencies that both my lawyer and I had overlooked—genuine issues requiring correction.

Portfolio Scanner: The AI read my investment portfolio, summarized holdings, and researched published analyst papers before rendering opinions on specific equities. Solid preparation for financial advisor meetings.

Literary Companion: It recognized every book I randomly mentioned and provided summaries in exactly the length and style I specified. For a book discussion group, it delivered chapter-by-chapter breakdowns plus discussion points—we nominated it as a virtual member.

The Biography Project: This was the ultimate test. A friend had dictated 50 pages of life story—unpunctuated dictation with some half-finished thoughts. Normally, this would require months of careful editing. Here’s what AI accomplished in hours:

• Instantly cleaned up punctuation and grammar across the entire manuscript. Two seconds of “think time.”
• Identified natural breaks and created logical chapter titles in three seconds
• Provided readability analysis. It chose the Flesch-Kincaid scale to do that.
• I asked for a critique of the document with suggestions for improvement and used several different AI engines to compare their “personalities.”.    
• Fact-checked dates and historical references
• Produced a polished, flowing 50-page biography in twenty seconds while preserving the author’s voice

The author’s role shifted from “editor with red pen” to “director with discerning eye.”

The Takeaways: Human-like Capabilities and Serious Concerns

Human-like Capabilities

• Totally “gets” context when provided enough description
• Mind-reading responsiveness to tone and style refinement
• Lightning-fast analysis of large datasets to identify patterns humans miss
• Ability to explain complex topics at any comprehension level and any length specified

Serious Concerns

• “Hallucinations” – Very rarely, it confidently and convincingly presented fabricated facts. It ‘fessed up when called out.
• Lacks social context and common sense that humans take for granted
• Still needs human oversight for direction and ethical guidance
• Because its knowledge and power is so impressive, it can create over-trust where outputs are accepted uncritically

The Bigger Picture

Harvard Business Review identified a total of 100 use cases for AI, all falling under the definition of “large language models” (LLMs). This represents one side of AI. The non-LLM side consists of things like robots and medical imaging.

There are hundreds of AI models designed to serve special applications and niche markets. For example, Thomson Reuters, after testing it in 1,200 law firms, just launched their AI system that compresses 10 hours of legal work into 10 minutes.

We’re experiencing infrastructure-level transformation comparable to electricity or the early internet. Current systems demonstrate PhD-level intelligence across most professions. Government attention is massive, AI investments are double that of any other sectors and growth follows 1990s internet adoption patterns—but much faster.

Scientists envision smarter medicine, household robots, faster drug discovery, and universal knowledge access. Forecasters say companies will either utilize AI or go out of business.

But concerns loom: data privacy, job displacement, misinformation, deepfakes, and existential risks from unchecked AI development.

Getting Started

Want to try it? Download ChatGPT and place it next to your Google Search icon. Start with the free version, then consider the $20/month upgrade. Alternatives include Gemini, Claude, Perplexity, and Grok. Unless you experiment with files and advanced “prompting,” the learning curve is close to zero. Not sure what to try? Just ask it!

This exploration continues. We’re all learning together as we navigate this remarkable technological shift. These tools don’t replace our brains (yet!), but dramatically expand our capabilities—think intern, not master.

Talk to us here or at gpk111@hotmail.com. Content reviewed by humans!

Six Fundamental Questions

Here are 6 areas to explore, including a few observations for each and some exploration suggestions in italics.

1. What Is AI Anyway? Unlike regular programs, AI learns from millions of examples, using 10x-1,000x more computing power than Google searches.            
   • Neural network concepts are at the heart of this, while the capital expenditures of big tech are staggering.
2. How Big Is It? It’s really BIG! This is transformative infrastructure with massive government attention and investments double that of any other sector. Analysts warn that soaring AI capital spending may outpace future revenues, raising fears of a bubble.”
   • Compare historical adoption curves, monthly active users (MAU), and software offerings to grasp the true scope of possible job displacements and great new opportunities.
3. Where Is This Going? From invisible daily infiltration to concerns about power concentration, job displacement, and existential risks.                                             
   • Peter Diamandis describes five major trends: ubiquitous communications, exponential AI growth, practical robotics, autonomous transport, and energy abundance—and potentially a billion robots within 25 years.
4. Where Can I Get One? Most engines are free or $20/month. Start simple, then  progress to specialized versions or even “AI agents.”                                                                                  
   • Voice assistants like Siri and Alexa are simple entry points, but are rapidly being merged with the direct chatbots like ChatGPT, Gemini, Grok, Claude and Perplexity.
5. How Can AI Fit Into My Life? From behind-the-scenes personalization like Netflix movie suggestions to more purposeful chatbot usage, AI is infiltrating our lives.                                                     
   • The Harvard Business Review article grouped the use cases into six categories, with uses “as wide-ranging as the problems we encounter in our lives.”
6. Any Additional Risk for Me? AI amplifies digital surveillance and gives scammers more powerful tools. Traditional safety measures like password changes are necessary but insufficient.                     
   • We already live in a “surveillance society” (see “Supervision: An Introduction to the Surveillance Society” by Gilliom and Monahan) and AI amps that up—think buying suggestion pop-ups that seem to read your mind. Develop a list of personal security measures that reflects the AI era.  

Additional areas worth exploring: the environmental impact of massive computing requirements, social impacts like teenagers trading in their real friendships for chatbots, military weaponization and geopolitics, government regulations and guardrails, and the ultimate questions around superintelligence and technological singularity, where AI outsmarts humans.

Select References and Additional Reading

1. Harvard Business Review – AI Use Cases Studies

• Original Study (2024): “How People Are Really Using GenAI” – Study identifying 100 categories of AI use cases divided into six top-level themes: Technical Assistance & Troubleshooting (23%), Content Creation & Editing (22%), Personal & Professional Support (17%), Learning & Education (15%), and Creativity
• Published: March 19, 2024
• URL: https://hbr.org/2024/03/how-people-are-really-using-genai
• 2025 Follow-up: “How People Are Really Using Gen AI in 2025” – Updated research showing evolution of AI use cases, with new categories emerging including significant growth in companionship and therapy applications
• Published: April 9, 2025
• URL: https://hbr.org/2025/04/how-people-are-really-using-gen-ai-in-2025

2. Gilliom, John and Torin Monahan. “SuperVision: An Introduction to the Surveillance Society”

• Publisher: University of Chicago Press, 2013
• Pages: 178pp
• ISBN: 978-0-226-92444-1 (paperback), 978-0-226-92443-4 (hardcover)
• Book Summary: Comprehensive introduction to surveillance studies examining how everyday technologies (credit cards, cell phones, search engines, GPS devices, Facebook) create a surveillance matrix that monitors and assesses citizens, often imperceptibly. The book explores who uses surveillance, how it’s implemented, and its effects on society.
• Key Theme: “We live in a surveillance society” – demonstrates how routine daily activities connect people to surveillance systems
• Publisher URL: https://press.uchicago.edu/ucp/books/book/chicago/S/bo14365311.html
• Available: Academic libraries, Amazon, Google Books, Barnes & Noble

3. Peter Diamandis – Five Technology Meta-Trends

• Specific Source: RISKWORLD 2024 keynote presentation, San Diego, May 6, 2024
• Five Meta-Trends Identified: (1) Ubiquitous Communications and Sensors, (2) The Growth of AI, (3) Robots are Coming, (4) Future of Autonomous Transport, and (5) Energy Abundance
• Key Predictions: “In 25 years, there could be a billion bipedal robots” (quoting Vinod Khosla), radically changing GDP, productivity and human happiness
• Market Context: Part of broader predictions about exponential technology convergence, includes forecasts about AI achieving human-level intelligence, and references to “Abundance” and “Singularity” frameworks
• Documentation: Risk Management Magazine interview, May 7, 2024
• URL: https://www.rmmagazine.com/articles/article/2024/05/07/riskworld-2024-peter-diamandis-shares-five-trends-driving-future-innovation
• Additional Resources: Peter Diamandis official site: https://www.diamandis.com