12 Most Misused Buzzwords In It

Every year or so the world gets new tech speak to master.

The smartphone, the cloud, augmented reality, the metaverse, containerization, and bots are just some of the techie words that have entered the vernacular in the past two decades.

Sometimes a new term is easy to grasp and clearly defined. The term “smartphone” is case in point. You’d be hard-pressed to find someone who wouldn’t recognize it for the miniature computer that it is, or someone who would confuse one with a flip phone or rotary-dial.

Other terms are harder to get. Want proof? Ask some colleagues outside of tech to define a digital twin or describe the difference between virtual reality and the metaverse. Good luck with that.

To be fair, the technology world doesn’t make it easy. Tech terms aren’t usually intuitive (who came up with “the cloud”?), and some terms have varying definitions even among technologists.

Those dynamics have led to a lot of misapplied, misused, and misunderstood terms, particularly in business environments where many professionals are nervous about being left behind so they talk about the latest and greatest — even when they’re don’t understand it — to show they’re keeping up.

Some technologists do the same.

As a result, people co-opt tech phrases, fail to get the terminology right, and overuse jargon — often to the point where the terms begin to lose their meaning, breaking down communication, collaboration, and, at times, business-IT alignment.

CIO.com periodically polls tech leaders to spell out the terms they most often hear misused. Here’s what they have to say about the most misused buzzwords today.

1. Digital transformation

“Digital transformation” continuously tops this list, cited by more sources over multiple years than any term. That shouldn’t surprise anyone in IT — or business in general — as nearly any change (big or small) seems to be labeled transformative.

“For ‘digital transformation,’ it has so many meanings and is mostly a marketing ploy — and a very clever one at that. The term has been used by just about every vendor to market a product, and the term did create a lot of hype in the technology industry. There is no meat behind the words, it can mean almost anything that marketing needs to sell,” says Sue Bergamo, CIO and CISO at BTE Partners.

Others had similar opinions on this ubiquitous catchphrase.

“Everyone talks about digital transformation, but too often they’re using it to simply describe digitalizing a paper process rather than fundamentally changing how the business operates, which is the transformation part,” says Mike Trkay, CIO and chief customer officer at FICO.

He sees that misconception as a problem.

“There are moments when the business owner will talk about digital transformation and they just view it as a change of technology. But it’s a lost opportunity if they’re not taking a chance to really look at those processes and think about how they’re running the business and how to change to be more efficient, to drive better outcomes, and drive more value,” he says.

2. Artificial intelligence, machine learning, and intelligence in general

Tied for the second spot are all the terms around artificial intelligence: AI, machine learning, and just about everything else around digital intelligence. Technology leaders say there’s general confusion about what these terms actually mean, with many now using “AI” to denote anything having to do with automation or basic algorithms.

Using the term to incorrectly encompass such a broad spectrum of algorithmic technologies seems to downplay true AI’s transformative power.

“My absolute favorite misused buzzword these days is ‘AI.’ AI is everywhere and everyone wants in on the game,” Bergamo says. “Artificial intelligence has been around for many years and is finally earning the stripes that it deserves. I highly doubt that the world will actually be taken over by a bot, but one thing is certain: True AI can perform so many meaningful tasks that can help everyone do something just a little bit better.”

3. Responsible AI

There are some AI-related terms that also make this list.

One is “responsible AI,” which Trkay says is often too narrowly construed — a mistake that could come back to haunt organizations.

“Too many people limit the idea of responsible AI to data protection, and there are so many other critical parameters to responsible AI — explainable AI, auditable AI, ethical AI,” Trkay says. “I feel like too often when I hear people talking about responsible AI, they’re only talking about data protection and data leakage. Don’t get me wrong. Those are important. But we’re creating some real exposures if we’re not addressing biases or hallucinations or the explainability or auditability of it.”

4. Agentic AI and agents

Doug Gilbert, CIO and chief digital officer at Sutherland, says many people treat the terms agentic AI and AI agents as if these technologies have pushed out all other forms of AI.

“I’m being asked, ‘Why am I going to do generative AI if I’m going to do agentic?’” Gilbert says, noting that he often finds himself “giving a reality check on what each of these things mean.”

Jamie Smith, CIO at the University of Phoenix, concurs.

“An emerging buzzword that is misunderstood or misused is ‘agent’ or ‘agentic.’ It seems to be most often confused with virtual chat assistants or any AI that conducts an action or task,” Smith says. “At the core of true agentic systems today are the notions of autonomy and goal-seeking behavior. Agentic systems gain autonomy as trust increases and improves outcomes by being given goals at which to optimize. Without these characteristics, they aren’t really agentic but merely AI workers.”

5. AI hallucination

Gilbert sees a similar trend with the term “hallucination” in that it’s often incorrectly applied to all issues with AI.

“Any error in the world of AI is attributed to hallucinations versus understanding or acknowledging that it may be problems with the data or the model or not formulating prompts accurately or having biased data,” he says.

6. Large language models

This is a point of semantics for Mark Sherwood, executive vice president and CIO at Wolters Kluwer.

“Everybody gets the fact that they’re large. And every year they’re getting larger. So let’s just call them language models. Just call them LMs,” he says, admitting the use of “large” here is just a pet peeve of his.

It may only be a pet peeve for Sherwood, but the question of what model gets to identify as large may have increasingly important business implications as organizations consider how large they require a language model to be to meet their needs, whether they’d be better off using small language models for some work, and who sets the parameters for what language models are labeled “large” or “small.”

7. Zero trust

Here’s another tech term called out by Sherwood for being something of a misnomer.

“It makes it sound like nothing is trusted, which isn’t actually accurate,” he says. “It’s not zero trust; it’s like specified trust. It’s just that we’ve chosen who to trust. There’s still a level of trust.”

Bergamo also cited “zero trust” as a problematic term, but for a slightly different reason, noting that the term overpromises what it actually can deliver.

“Zero trust is a concept; it is not an architecture or a product — though there are products that support zero trust,” she says. “Every time a company implements a product, think multifactor authentication, a portion of the zero trust concept is enacted, but does not guarantee that the overall environment is secure. True security to ‘verify, then trust’ starts with a layered approach, where security products and practices are implemented to further harden an environment.”

8. Tech debt

The term “tech debt” continues to plague the IT department nearly as much as the actual tech debt itself.

The issue is that the term that can mean different things to different people, inside and outside of the IT department. Some define it as problematic code knowingly deployed for the sake of speed, with the understanding that teams would fix it later. Others use the term to refer to legacy systems or the cost of maintaining them.

“In theory, it represents the accumulated cost of deferred maintenance or architectural shortcuts,” says Karen Swift, founder of Cybellis Consulting and former vice president of IT at Penske Media. “In reality, it often masks systemic underfunding or defunding of core systems. As organizations accelerated moves to cloud and SaaS architectures starting around 2020, many legacy systems were deprioritized, creating real operational and security risk. ‘Tech debt’ is a more palatable phrase than admitting systems are outdated or brittle, and it is most often used by executive leadership and finance teams rather than practitioners.”

9. Innovation

Leave it to modern business to misuse a longstanding, well-defined word.

“Innovation is frequently framed as experimentation without a clearly defined path to accountability, scale, or measurable outcomes,” Swift says. “This is especially true when paired with phrases like ‘AI innovation,’ which has become a broad umbrella for nearly any AI-related activity.”

10. Automation

Swift says it’s not just “innovation” that’s incorrectly used. She says “automation” gets similar treatment.

“‘Automation’ is often used to describe scripting or RPA efforts, but without sufficient attention to end-to-end process redesign, change management, or sustained efficiency gains,” she explains. “Automation without rethinking the business process simply accelerates existing inefficiencies.”

11. Post-quantum computing

As he does with the term “large language models,” Sherwood has a quibble to the first word here — the “post” part of the phrase.

“‘Post-quantum computing’ — it makes it sound like an event; it’s like Y2K,” he says. “‘Post-quantum computing’ makes it sound like something happens on Day X when it’s actually going to be a slow drip. There will be post-quantum computing, but I don’t know when to draw that line. Is it when everyone has access to it? When some do? It’s just going to be quantum computing, I don’t think it comes in with a bang. It will be gradual, and it will be hard to define when it’s here.”

12. The business

This one hits close to home, and it goes to how IT may view itself and how the rest of any given organization may view IT.

“While it may not be a true buzzword, one of the most misused terms is ‘the business’ or using the term ‘customer’ and ‘the business’ synonymously,” Smith says. “Thinking about business and technology as divisible entities at this point in the game is getting it all wrong. Thinking about ‘the business’ as a customer and technology teams as consultants to that customer creates a level of abstraction that slows the ability to impact the true end customer/consumer with digital experiences and solutions. Technology and business are inextricable partners in driving innovation, efficiency and outcomes, especially with advancements in artificial intelligence.”