Garbage In, Garbage Out: Why Some People Suck at Using AI in the Workplace

The appearance of AI in the workplace is making a huge impact in the day-to-day workflow of professionals across highly varied industries.  But have you noticed some professionals are really ‘bad at AI’, if there could be such a label?  We have, and the core concept of garbage-in, garbage-out is exactly why some people deliver low-quality work even with AI power now on their side.

Some professionals are ‘bad’ at using AI productivity tools

Appropriately, the computer science principle of ‘garbage in, garbage out’ (GIGO) still rings true with our application of commands to prompt our AI tools to do what we need them to at work.  Though this space is ever-evolving, there is a core workflow principle at play here too that we just can’t ignore. 

Sloppy work starts with sloppy thinking and prioritizing the wrong initiatives when solving a problem.  Core principles, best practices, and strategic planning are emphasized in business for this reason.  When you start on the wrong path, it’s hard to course-correct. 

Poor work product typically begins with a lack of discipline and sub-par decision-making skills that can ripple down to even what AI-tools create for these individuals.

Of course, we do not yet know what AI tech will be able to do in the future and if the tools we use will be able to fix poor human input entirely.  We CAN talk about the important topic of how we are using these tools right now as we transition into an ‘AI-in-the-workplace-and-soon-every-area-of-your-life’ perspective, though.

How to prompt AI effectively with a classical approach

The solution for poor work output through AI is the same secret to success that has always worked for problem solvers and workflow for businesses across decades…and history.  Half the output we get from AI tools lies within the prompts they are given by the user.

This concept is based on a classic quote attributed to Socrates that says, “understanding a question is half an answer.”  This idea also ties into the Socratic Method and Stoic principles of philosophy. 

This sage advice and insight still applies today. It is a classical concept of learning and improvement we can continue to explore as professionals and goal-oriented humans, no matter how we evolve our environment.

AI is a computer.  If something is going wrong, it’s most likely the human half of the equation missing the mark.**  Therefore, it is up to the AI-tool-operator to determine when a goal is fulfilled and how far to pursue it through AI tools for their final product.

**Yes, AI can make mistakes.  It cannot decide to actually apply the output it generates in most workplaces without our help, though.  As of today, a human decides whether to use the ugly AI flyer ChatGPT spit out on the first try, or to dig deeper and refine.  We’re talking AI tools versus AI systems that are automated.

So how does this quote from an ancient philosopher help with getting the best results from AI in the workplace?

AI prompting best practices meet Socratic thinking skills

When you consider the commands you are inputting to an AI-tool to get your desired research, implement a workflow solution, or generate a creative asset, it takes a bit of thought and practice to learn how comprehensive you need to be – or CAN be – with AI search.

You must be specific about any visual details that are important to you with the creation of an image.  For research, the depth and specificity of your question will dictate the level of detail you receive back in the results from AI.

When AI gives you exactly what you requested back and covers all the bases, it’s a success and feels like magic!   The information meets your needs and points you in the right direction towards your goal.

Oppositely, we ‘miss’ at applying AI successfully in the workplace when the output from a tool falls short on fulfilling the original need we had in mind. 

We’ve all been there before, too.  You give ChatGPT a prompt you think is complete, but miss a key detail that will modify the entire answer.  Or if you forget to include a key design-element of a graphic you are creating in your description, it will not be included by AI in the output.

We hold the power with directing our AI-tools in the workplace. Our questions have a significant influence on the AI-answer, and applying our analytical ability to dig deeper and modify the output leads to the same results you expect in a conversation, or interaction.

Professionals who are curious and work towards the highest standards in any setting often repeat this habit with utilizing AI-tools in the workplace too.

Importance of ‘the question’ in your AI workflow

The type of prompting popular AI models like Gemini and Claude use are similar in logic and search flow to the way Google Organic Search works, but AI takes answering a question several steps further.  Searching for an answer through AI allows you to qualify your question within the inquiry itself and continuously revise.

This means AI-tools are capable of taking on large requests in the realm of research, proofing, and creating unique content based on user-set parameters. The possibilities and power are endless…even scary at times.

The caveat for harnessing AI-power effectively is that a user must really know the nuances of their question before asking it, or at least have a detailed description of the problem they are trying to solve in mind.

Without fully understanding what end-result is desired, there is no way to direct an AI tool to generate it. Even the task of brainstorming with AI requires seed-parameters to start the process and define the action of ‘brainstorming’ effectively.

The beauty of problem solving actually happens when you fully understand your need, or question at hand, and this concept relates back to the Socratic Method and principles of constructive questioning.  AI prompting best practices tie into this concept and skill of analysis directly.

How ‘visualization’ benefits AI content generation

As humans, we are taken one step closer to our solution through considering our need, even before AI helps us out.  Visualization takes us at least half-way to the end result we want, because it is impossible to reach that destination without ‘seeing’ it first.

If we tap into our inner student and allow our human problem solving skills to take the lead, this practice of ‘starting any challenge by considering the desired solution’ can create tremendous success – especially when combined with the speed and power of AI technology.

Your human vision is a powerful discernment tool in guiding the end-application of AI.

The art of editing within AI workflow optimization

Remember the concept of GIGO is always in play as you interact with AI-tools and should be a central concept of focus in your AI workflow optimization.  Whatever you prompt the tool to create is what you receive back.  Garbage begets garbage.

The human who is driving an AI tool must first critically review its results to verify if this asset is useable as-is, or if they need to go deeper into modifying the request to get a more accurate output.

An example I personally experience all the time in my own use of AI for work: SEO keyword research.  I use ChatGPT and various models to help with this task routinely. There are a number of different reasons I could be looking for keywords, like writing website copy, or writing a blog.  In any application, I’m finding the best terms to focus on for SEO-purposes through AI-tools.

If I accept the first round of results that a Claude, or Nano search gives me, I find that I often missed one parameter I need in my prompt – like asking to include TOFU, MOFU, and BOFU categorization that can be pulled into the results table for blogging, or estimated keyword volume in the US and UK.

Or I may have been so specific with my command for AI that I excluded related keyword variations by accident.  In that case, I would need to modify my request with AI to get the full results and high traffic keyword phrases I was hoping to see, or start over with my request.

As we use AI in the workplace and find ways to maximize our time and workflow efficiency with these tools, it is important to remember the principle of continuously questioning for deeper understanding. ‘Persistence in getting the best results’ is a positive professional trait that stands the test of time. 

Think critically, and keep editing until you get the data, info, or creative, you originally envisioned – or an even better end product than you first visualized.

Helpful AI workflow prompts to dig deeper and edit output:

• “Explain [concept] further”  when you receive AI explanations with concepts, acronyms, and phrases you do not understand.

•  “What data supports [x] claim?” when you receive information that needs support from a credible source, data to fully understand the AI explanation, or that does not seem credible.

• “What if [x] is true?” to get further enhancements to the original AI answer based on information you recall, or find, that could impact the results.

How ‘talking to AI’ equals business process optimization

Very much like human interaction, Socratic-style questions and deep-dives with AI work like a conversation.  This makes sense with the way AI models communicate and reason.  Without getting into the technical side (at all), the user experience is more conversational than targeted Google Search results.

Google bases its results on keyword relevance to your original query and giving you a ranked list of pages with that answer.  So you are receiving a more “exact match to less exact match” set of answers that connect to your keyword.

Even in concept, this experience is personalized, but from a distance.  Google can get eerily accurate too with results like it is reading your mind.  But with AI, it’s more like having a genuine interaction than an interactive encyclopedia of the web at work.

What we have found in our day-to-day workflow as professionals is that ‘talking to AI’ and ‘learning from AI’ are real systems and exchanges that take place and create progress with an initiative. Their success is largely due to the real-life workflow users experience with these tools. It feels like a conversation with the most helpful implementer…ever.

How does successful use of AI in the workplace look?

This interaction when using AI-tools is a largely positive habit in the workplace when the goal is ‘refinement’ and reaching the best result on a deliverable. Getting clearer data, engaging in fine edits of creative assets, and asking for deeper explanations, or revisions are the same habits the best professionals in the workplace already use.

The folks who are putting in time to monitor their AI results like a real tool are communicating to get closer to their goal.  They cultivate the output they receive and get a better product through AI assistance than those who are relying on AI like a surrogate for their own thought power.

These driven professionals are also not accepting AI-results alone.  They are supplementing with further human refinement.

For example, we all know a professional who uses ChatGPT to help with work writing. The stand-out team members in this group will do a further personal proof and edits before finalizing the piece.

Why are some people are ‘bad’ at using AI tools?

There is a fine line between training an AI-tool to assist efficiently with your work and hoping you can delegate your entire workload to an AI tool, though.  The attempts we have seen from the average office worker through executives to teach an AI to do their work for them have ended in disappointment as of 2026. 

There is also some moment of outrage with Claude, that virtual scoundrel. Regardless of your level of AI skill – good, or bad – we have all battled him at some point.

Alas, there is a limit to ‘Claude-conditioning’ you can do when out in the real workplace as of today.  AI is a wonderful tool that helps us do our jobs better, but GIGO still applies as a principle.

Despite this fact, many real professionals use AI to assist with their work…and you can really tell. 

How many times have you seen an ad, a written piece, or even a LinkedIn message that is clearly AI generated, and you say, “ugh, this is AI” in disgust?  We all have, and it’s a new, but pointed critique. 

Whenever the use of AI is pointed out, the consumer of that piece, idea, or product is expressing disappointment in the depersonalization and clearly lazy use of AI tools to produce an asset.  The content feels insincere and thrown together.

Case in point.  If this clearly-AI-generated product is from a business, the leaders and owner should be keenly aware – it translates SO poorly to their audience.  What the consumer sees is that your company puts shallow effort into their forward-facing messaging, or products, by leaning on AI to do the work that requires actual thought-power.

This negative perception can easily snowball when other shortcuts and misses are detected by your customer base too, and it’s a tough image-association to out-run.

This is a big danger to businesses in the age of AI.  When you have any level of team member misusing these AI-tools without the integrity expected for a business product, or leaning too heavily on them to replace thought power behind their work, they harm your brand.

The same, under-performing colleagues – now powered by AI

Culture, technology, and an infinite number of environmental and operational changes happen for businesses and the world over time.  We will always find the need to evolve as professionals for this reason, but the core skills that make one worker a total badass and make the next a waste of space remain true throughout history.

The worker who is looking for the shortcuts, the easy way, and how to do as little actual work as possible has always been around, and you know them.  As a business owner, you may detect these individuals on your team when it’s too late to easily remove them. Or as an entry-level team member, you may be reporting to this jack-ass.

Wherever they fall in the org chart, these individuals with poor work ethic have not changed, they are just powered by AI now.  The danger in these tools for the lazy pro are that they enable shallow work that can often pass as ‘complete’ and unique enough to not worry about being caught for direct plagiarism.   

The vast majority of school systems have already put parameters in place to prevent students from using AI to avoid actually completing their work.  This type of oversight is critical, because we need to ensure learning takes place for K-12 students that embraces the new world we are in and becoming AI-literate.

Why is this same standard schools set out not being followed with AI in the workplace?

How our humanity influences AI output

Different personalities that exist at all levels of a business-hierarchy approach their work from various motivations – some good, some indifferent, and some nefarious.  The central theme is the fact that we are all very human, and our emotional and process-based thinking is part of our real work day-to-day. 

Our mortal limitations alone will keep us from using AI tools perfectly.  Our natural problem-solving states are directly connected to the way we interact with these tools and apply them.  We see evidence of this in the world every day.

The same people who stole work from others in the past, or who left their work to a team member to complete while they took the credit can now use these AI tools in a more sly and undetected way – or so they think.

Just as we described the feeling and lack of effort we have all experienced in seeing content that is produced poorly through AI, we can pretty much uncover who allowed that crap out into the universe as a business asset.  They are the same people who were always bad workers, and AI will not make them better performers as of today.

Poor AI workflow habits for businesses

A business utilizing AI in the workplace on any level should have a clear AI policy that everyone in the organization upholds – just like a school.  This information should also be available to consumers, or clients, so they are aware of what portion of your work is created, or contributed to, via AI.

But as of now, workers and organizations who are using AI tools poorly have a few hallmark errors in common.

• AI prompts are not detailed and omit key parameters needed for the most detailed, accurate, or attractive result.

• For writing, AI does the heavy lifting with little-to-no revision of the content by the human-operator afterwards.

• For image generation, AI composes the image, but the operator does not critically review the result for aesthetic harmony.

• For data, AI aggregates info for you, but it is not sorted, or organized in a useable manner.

• For research, the AI operator fails to gain substantial and complete information by further qualifying their search inquiries through the tool based on the results received. 

In all 5 of these examples, the AI operator fails to complete their part of the ‘conversation’ with their AI tool to generate the best results.  The problem-solving skills that have always been missing with individuals who are not the best workers are still manifesting.

Constructive AI workflow habits for professionals

On the opposite side of the spectrum, you may be wondering what AI processes, or habits, are green-flags in the workplace and translate well to your business’s objectives and audience.

Just as the bad habits and poor problem-solving skills some workers naturally apply have remained the same over time, the positive and productive skills you would expect to see are applied to AI tools and the command / refine process. 

These are a few ways the most effective professionals interact with AI-tools at work:

• Prompts are detailed and include the same considerations in the AI-queries that would be made for their own human-work. 

• They ‘dig deeper’ on answers to make sure any research, or creative assets are pristine and accurate. 

• They question the strength of the information generated by AI and continue prompting with deeper queries to uncover original sources and vet their information.

• There is no fear of ‘starting over’ with AI and trying a different query when needed to get the best results from the tool.

• Sharp quality control is employed to refine, or discard, work that does not meet quality standards. They always maintain oversight of the final product.

• They have a defined personal standard around how all tools – including AI – are applied to their work to preserve the integrity and strength of the company’s mission.

Of course, with this level of care, these pros get the best results from AI tools. You can’t tell AI was a part of the work at all – or how it was used. 

Having the help of AI technology at work along with the discretion to use it in an authentic way is the sweet spot for a business in our current times. The best and brightest professionals seamlessly integrate AI into their work, as they would with any tool. 

The company’s team and audience only sees quality output they expect…no matter if AI was used to create a final product, or not.

How AstroCat uses AI for operational efficiency

As an example from our AstroCat office, we have found success with leaning on AI-tools for research, especially in the marketing realm.  ChatGPT has been a godsend with uncovering hard-to-aggregate information that can be time-consuming to find, analyze, and organize.  We shave off hours of research time by employing AI.

However, we do not use AI for writing composition, outside of research (including SEO research for keywords), titles and headings of written pieces, and formatting.  It is important to us to define our written output as ‘original’ and as an important creative asset to our business. 

This is the reason we beg forgiveness on any typos you see in our pieces in our site’s AI policy. We do not edit with AI as of now and have no plan to implement this use as of this year.  Even though a final proof through AI could be beneficial to us to enhance our work, our belief is that presenting our product in urtext preserves the unquestionable authenticity of our written content and ideas as our own.

The integrity of our company’s AI-utilization and limitations to how we allow ourselves to enhance our work with these tools gives us, as Founders, peace of mind.   These values also contribute to our brand and forward facing company-image, but the internal confidence we maintain from knowing we produce outstanding work in our most important medium without relying on AI is a gift in itself.   

A musical perspective on authenticity and AI

You see, as musicians, the originality of our work and creativity we apply to the art of business are part of the ethics we developed with our musical craftsmanship.  The urtext term is from our classical music world to distinguish sheet music, or scores, that are the original text written by the composer, exclusively.  This sheet music has not been edited by a third-person – it is just what the composer conceptualized.

When applied to AI, this concept of delivering urtext for key assets combined with our musical backgrounds makes more sense.  To present work output that is tastelessly automated, sub-par, or practically stolen is wounding to our egos in any medium. 

Instead, every forward-facing asset we share with the world is a performance of sorts.  Even if our company, AstroCat Consulting, is creating a flyer for a party, that one item represents who we are and showcases our style, aesthetic, and abilities. 

With this performance-level importance given to a company’s content and communications, it becomes easier for workers and leaders to also prioritize authenticity – whether AI is involved, or not.

The ‘fix’ for garbage in, garbage out AI

Unfortunately, there is no cure for a bad personality, or values.  If you have a Machiavellian co-worker who uses AI tech to out-perform you on tasks that require speed, but there is no quality control established, you may feel tempted to also take their AI-shortcut.

We urge you not to cut these corners unless a shortcut is really just an efficiency-hack and no quality is lost on the actual work product.  An AI-shortcut is not worth engaging when you must sacrifice something of value to you as a professional, or to your organization, to reach your end-result more easily.

Any sound business solution will bring mutual benefit to all parties, even on seemingly un-beneficial changes. 

Like in this non-AI example of symbiosis for a business when rolling out a change, the use of AI-assistance for workflow requires consideration of the big picture. The efficiency gained must have an all-around benefit to those receiving and producing the output. 

A balance is needed between AI saving you time versus the effectiveness of your final deliverables. Otherwise, an application of this technology may not prove to be worthwhile for every use available.

Effective AI utilization for small business workflow

For professionals and leaders –  we encourage you to consider the ways AI saves you time and can enhance your operation simultaneously as you create your AI policies.  Remember, benefits must be symbiotic between company and customer, or client, and these rules should be decided in advance, not as you go.

You can always update your AI policies and best practices as times change and more-refined AI technology enters the workplace.

But in the meantime, how do you avoid the fall-out from the workers who are approaching the AI revolution like it’s the wild wild west and they may never need to think about work again?

Just as the Socratic thinking and questioning of AI output is important as a professional is working with these tools, leaders and managers at every level of a company’s infrastructure – large or small – should have similar systems of quality control in place.

Any content, creative, or product that is facing forward to your audience matters for business owners.  Maintaining the strength of your brand when it is so easy to take shortcuts with AI in the workplace is a new challenge and one that requires tuned-in oversight.

The great news is that the same amazing team members you always had can continue to thrive, even in changing times.  The key is embracing new developments, but also being pragmatic in how you integrate new tech into your workflow. 

A popular AI strategy for businesses that failed

To conclude, I leave you with my very on AI-story as a consumer that may make you rethink some of your automated communication tools, like Chatbot if you run a service business –

Which was worse – the mice, or the chatbots?

Consequences of poor AI-implementation for businesses

There are a ton of take-aways from this story and the consumer experience of running into chatbot after chatbot in an emergency.  For our purpose and considering what doesn’t work for everyone in the AI-realm right now, this is one of those tools that gives a perfect example.

Many companies and pros want these chatbots to do the mindless work for them and serve a specific purpose of keeping customers engaged…in the sales funnel. But so few companies using them actually think about the customer perception and risks around rolling out this type of communication tool without a lot of careful planning and testing first.

This is the danger of garbage in, garbage out AI on the enterprise level and when the wrong, careless choices on the part of the AI-operator actually lower the standards of a business.

Rather than using this sophisticated AI-technology to vet a new client and keep them on the hook when they are a small business and unavailable, they could more-easily choose the lesser function of setting a dynamic away-message.

This type of AI tailoring for requests gives a clear picture of what to expect with the next communication, rather than forcing customers to chat with AI as a gatekeeper…and just because every other service is using it.

If you must program a chatbot to ask vetting questions of your leads, or customers, focus on answers to THEIR questions, if they want the answers right then. Also, limiting any questions asked of them ahead of speaking with a human, outside of call-back info.

This prioritization of the human aspect of your business and tasteful application of AI is what will set your company apart from your competition. It will serve as a powerful differentiator in an ever-changing workplace. 

AI in the workplace that WORKS – in conclusion

If you are a professional utilizing AI in the workplace and prioritize excellence over shortcuts, your consistently-high quality of work in your deliverables will still shine through. The AI-driven surge in workplace productivity will balance out, and shallow work will always reveal itself in time.

Stay the course, and you will know when the time and tech are right to embrace the next AI-development for your business.