What Grok Said about AI

What Grok Said about AI

Recently, I asked Grok about the use of AI in fields other than writing.  Here is my question, followed by Grok’s response: 

My Question:

As a writer, I understand the value of AI. For example, I use it as a first draft creator and as an aid in research. So I see how it saves me time. What I keep hearing is that AI helps in all areas. I don’t understand how it helps physicians, plumbers, lawyers (except as the research tool), technology like self-driving cars, etc. Can you explain how AI is used in fields other than writing?

Grok’s Response

AI is transforming nearly every profession by handling repetitive, data-heavy, or pattern-recognition tasks that humans find time-consuming or error-prone. Just as it generates first drafts and researches topics for writers, AI acts as a tireless assistant, analyst, or co-pilot elsewhere—freeing people to focus on judgment, creativity, relationships, and complex problem-solving. It doesn’t replace humans in most cases; it augments them by processing vast amounts of data faster and more consistently than any person could.

Here are concrete examples from the fields you mentioned (and a couple of others for breadth), based on real-world deployments as of 2025–2026.

Physicians and Healthcare

Doctors spend huge portions of their day on documentation, reviewing records, and routine analysis—tasks AI now automates or accelerates dramatically.

Ambient AI scribes listen to patient conversations (with permission), automatically draft clinical notes, summarize histories, and even respond to patient messages. This cuts after-hours charting time (some physicians save ~2.6 hours per week) and reduces burnout by 20–30% in studies from hospitals like Mass General Brigham, Emory, and Kaiser Permanente. Doctors report they can actually look patients in the eye instead of typing on a computer.

Clinical decision support and diagnostics: AI synthesizes patient data, symptoms, lab results, and the latest research in seconds to suggest diagnoses, flag risks, or spot issues in medical images (e.g., detecting bone fractures that urgent-care doctors sometimes miss). Tools like AI co-pilots or systems such as ArteraAI provide risk scores for treatment decisions in oncology or cardiology.

Predictive and personalized care: AI analyzes electronic health records (EHRs) for conversational search (“Show me this patient’s history related to abdominal pain”) or predicts complications, enabling earlier interventions and tailored treatments.

The net result: more time with patients, fewer errors, and less administrative drudgery—very similar to how AI drafts save writers time.

Lawyers (Beyond Research)

Legal work involves mountains of documents, contracts, and data review. AI handles the drudgery so attorneys can focus on strategy and client advice.

Document review and e-discovery: Generative AI scans thousands of emails, contracts, or case files in minutes, summarizes them, flags “hot” documents, spots patterns, and even prepares deposition outlines or timelines. What once took junior associates days now takes minutes.

Contract analysis and drafting: Tools review clause language against market standards, highlight deviations or risks, and generate initial drafts of motions, briefs, leases, or settlement agreements. Some systems let lawyers slide a control to make clauses more favorable to one side.

Predictive analytics: AI assesses the likelihood of winning a motion, settling vs. litigating, or estimating case costs based on historical data—helping lawyers advise clients more accurately and efficiently.

Many firms now use these tools in pilots, reporting productivity gains of 100x in routine tasks while still requiring human oversight for accuracy and ethics.

Plumbers and Trades (Construction, Field Services)

This might seem surprising, but AI is already embedded in the tools and software plumbers use every day—mostly on the business and diagnostic side, not the wrench-turning itself.

Business operations and customer service: AI voice agents answer calls 24/7, triage emergencies, schedule appointments, provide instant quotes, and handle follow-ups or billing. Platforms like ServiceTitan, Jobber AI, or Housecall Pro use AI to optimize routing, dispatch technicians, generate estimates from job history, and automate invoicing—cutting admin time and increasing revenue.

Smart systems and on-the-job aids: Manufacturers build AI into plumbing hardware for leak detection (e.g., Moen Flo monitors water flow patterns and alerts homeowners—and plumbers—before damage occurs) and predictive maintenance. Field techs get AI troubleshooting assistants or apps that suggest fixes based on symptoms.

Estimating and marketing: AI analyzes drawings for material takeoffs, suggests upsells, or even helps target local leads with personalized marketing.

For a small plumbing company, this means fewer missed calls, faster jobs, less wasted inventory, and more predictable cash flow—again, saving time and reducing errors the way research aids do for writers.

Technology (Self-Driving Cars)

Autonomous vehicles are one of the most visible AI success stories because the entire system runs on AI.

Perception and decision-making: Cameras, LiDAR, radar, and sensors feed data to neural networks (a form of AI) that identify objects (pedestrians, cars, signs), predict their future movements, and plan safe paths in real time. The AI fuses all this sensor data, decides whether to brake, swerve, or accelerate, and controls the vehicle—far faster than a human could react.

Continuous improvement: Companies like Waymo and Tesla use machine learning to train on billions of miles of real and simulated driving data. AI simulates dangerous scenarios to test safety before real-world deployment.

End-to-end autonomy: Modern systems are shifting to “end-to-end” neural networks that go straight from raw sensor input to driving actions, handling everything from lane-keeping to complex urban navigation.

This is why self-driving cars feel “smart”—AI is literally the brain making split-second judgments.

Quick Look at a Couple More Fields (to Show the Pattern)

Manufacturing: AI predicts when machines will fail (preventing downtime), optimizes production lines in real time, and powers robots that adapt to changes.

Finance: AI detects fraud in milliseconds, automates trading, personalizes banking advice, and assesses loan risk more accurately than manual reviews.

In every case, AI excels at scale, speed, and pattern recognition—things that used to eat up professionals’ time or introduce human fatigue errors. The human role remains central: interpreting results, exercising judgment, building relationships, and handling edge cases. Just like a writer still shapes the final story after an AI draft, physicians, lawyers, plumbers, and engineers use AI outputs as a starting point or support tool.

The technology is still evolving (and improving rapidly), but the time-saving, error-reducing, and insight-generating benefits you already see in writing are playing out across the board. If you’d like deeper dives into any specific field or examples of tools professionals actually use, just let me know!

An Afterthought

Grok, along with Claude, Perplexity, ChatGPT, and Gemini, have become indispensable tools for me.  As I told Grok earlier, I use my AI buddies primarily for creating first drafts and research.  

While negative views float about of students using AI to write the essays that they are assigned to write for writing classes, and sites like Owlcation and LetterPile converting from using actually writers to AI driven material, it must be recognized that any new tool can be misused even as it is mostly well used.

There does then exist online tools that purport to detect whether a text is written by AI or is human created.  These tools can be fatally flawed.  For example, I have submitted many of my essays that I wrote well before I ever heard of AI, and some AI detectors will claim them to be mostly written by AI.  

Why do my essays appear to be AI created?  Because my grammar and organization are virtually flawless. I write like a professional writer because I have so much practice. (I taught English Composition for  over a decade at Ball State University, where I earned two MA degrees and a PhD., writing two masters’ theses and a doctoral dissertation.)

Interestingly, the same essay will be deemed AI generated by AIDetector.com and completely human created by Quillbot and phrasly.ai. So using the AI detectors requires some scrutiny.  

When AIDetector.com detects grammar/mechanics errors or typos in my text, I use it to go back to my text and repair the damage.  Interestingly, after I repair the errors and resubmit the essay, it now figures higher in its likelihood of being AI generated.  

Of all the detectors I have encountered, only AIDetector.com gives a few details regarding its conclusion.  Thus it can be helpful in spotting grammar/mechanics errors and typos.  

But if you are a careful, well-organized writer who seldom commits grammar/mechanic faux pas, keep in mind, that the better you write, the more likely you are to be taken for using AI generated text.

For example, I placed this “Afterthought” into AIDetector.com right after I first drafted it.  It was deemed “Likely human 12.75% AI detected, with the following analytical details:  

The text exhibits clear human markers such as grammatical errors (e.g., ‘earn two MA degrees’ instead of ‘earned’, ‘z AIDetector.com’), typos, and run-on sentences, which contradict AI’s typical flawless grammar and uniform structure. It includes personal anecdotes, lived experiences (teaching career, degrees), and ironic self-reflection about AI detectors flagging good human writing, adding a genuine, digressive tone absent in AI’s predictable, encyclopedic style. While some organization exists, the varied sentence lengths, informal phrasing, and absence of moral framing or seamless transitions strongly indicate human authorship.

Notice that grammar errors flag the writing as human, and helpfully, the analysis points out a couple of them. (Actually those errors are not really grammar; they are mechanics or typos.)

But the most egregious claim about this evaluation is that it states that there are “run-on sentences,” and there are no run-on sentences in this text.  It may be confusing compound and/or complex sentences with run-ons.  (There is a big difference.  Look it up!)  So using these AI tools can be very helpful if you stay alert to their pitfalls.  

Genuine writers who care about and are consciously invested in their writing skills know how to use AI as a tool.  They would not consider relying on AI to express what they genuinely think and feel.  

Just a the typewriter’s use over the fountain pen did not develop text expressing what is in the heart and mind of the writer, AI is simply a modern-day typewriter, making the mechanics of writing somewhat easier.

AI is useful in gathering information that already exists; it cannot be counted on to replace the unique information that resides in the human heart and mind of the individual.  Human beings can think and feel; AI can only search and gather. 

Don’t be fooled by fact that AI can string sentences together using what is has searched and gathered.  It still cannot express a heartfelt thesis—simply because it has no heart.  The human heart and mind cerated AI, but AI cannot create anything new.  That gift to humanity still comes only from one source—God.