How to Detect AI-Generated Content

AI-generated content is copy such as blogs, marketing materials, articles and product descriptions written by a machine. The AI content creator generates the content after a person inputs information such as keywords, phrases and topics.

AI-generated content can help speed up the writing process, and businesses are starting to take notice. Marketers may consider using AI-generated content for automating the content marketing process, which can, at times, be time-consuming and expensive.

How does AI-generated content work?

To produce content, users enter the text format needed, a topic and keywords. The AI then searches the internet and pieces together copy fitting the requested format. Some tools also rewrite a business’s existing content to add information based on web searches and editing tools. These tools can take this existing content and reformat it to fit other communication mediums, such as turning a blog article into a social media post or email draft.

For now, there are ways to figure out what’s computer generated

New research shows that people learn to spot machine-generated text.
Knowing what’s written by AI is getting more urgent with the proliferation of software like ChatGPT.
But some experts say that recognizing AI-generated text will become impossible as AI improves.

It’s getting harder and harder to tell the difference between sentences generated by artificial intelligence (AI) and people.

The good news is that a new paper shows that people can learn to spot the difference between AI-generated and human-written text. The ability to determine if computers craft online information is becoming critical with the rise of large-scale language models such as ChatGPT, which many predict will touch virtually every corner of our lives.

“If you receive text messages suggesting your grandmother is in urgent need of funds for a medical procedure, you’d like to know if it was a mass AI scam or a real family member,” Kentaro Toyama, a professor of community information at the University of Michigan, and author of “Geek Heresy: Rescuing Social Change from the Cult of Technology,” told Simmyideas in an email interview. “And, at moments when real human connection matters, we’d like to hear from real people, not machines.”

Telling the Difference Between Real and ChatGPT

Researchers from the University of Pennsylvania wrote their paper by examining data collected using “Real or Fake Text?”, a web-based training game. Participants are asked to indicate whether a machine has produced a given text in a yes-or-no fashion. This task involves classifying a text as real or fake and scoring responses as correct or incorrect. The study found that participants scored significantly better than random chance.

“Our method not only gamifies the task, making it more engaging, it also provides a more realistic context for training,” Liam Duga, one of the study’s authors, said in a news release. “Generated texts, like those produced by ChatGPT, begin with human-provided prompts.”

At moments when real human connection matters, we’d like to hear from real people, not machines.

Toyama said there are many reasons why learning to distinguish between computer-generated text and human writing is important. For example, teachers would like to know whether students are submitting essays they wrote themselves or text written partly or entirely by computer. He noted that recently, Vanderbilt University sent an email written by ChatGPT to its community linking the tragedy of the Michigan State shooting to campus efforts toward inclusion. There was a swift backlash.

“There are plenty of other such situations,” he added. “In fact, I believe that one essential form of regulation of AI is that the law should require any text, image, audio, video, or other creative output generated by a computer to be clearly marked as such.”

What are the pros of AI-generated content?

AI-generated content can be valuable in several ways. Here are some of the pros.

Efficiency and scalability

AI can create content much faster than people, which is probably the biggest benefit. An AI tool can produce an article in minutes. It would take a human writer much longer to do all the research and write it.

Multiply the quick turnaround by the number of articles, and an AI tool can produce a significant amount of content.

AI also helps with language localization for various geographic areas and can create social media with personalization for various sites.

Cost-effective

Hiring quality content writers typically costs a few hundred dollars per project, depending on the length of the article, the number of pieces and the needed technical knowledge. And this may be money well spent for high-quality, well-researched content.

Some AI writing tools are free, while others charge a monthly subscription rate. The pricing typically runs about $100 for tens of thousands of words.

AI-generated content may be better suited for simpler content than articles needing expertise and authority.

Improves SEO

AI content generators scroll through thousands of online documents to absorb the information. By viewing all these documents, generators choose keywords to improve search engine optimization (SEO). The AI tool can suggest keywords for the content writer. By using these keyword suggestions throughout an article, the content appears higher in search engine rankings if it follows the rest of the guidelines of being authoritative and is written by a person.

Overcome writer’s block

Writer’s block is a common hurdle for many people. At times, writers may have trouble creating authoritative content for a subject they know little about.

To help overcome this hurdle, AI tools can create detailed outlines and key points to help the writer determine what should be included in the article. AI tools can help the person overcome writer’s block and spark ideas to get started.

What are the cons of AI-generated content?

There are some considerations for AI-generated content. Some content is best written by a human writer. Here are some cons of AI-generated content.

Quality concerns and possible plagiarism

AI relies on data and algorithms for content. The intended tone may get lost. AI tools can cover black and white areas of a topic, but gray areas are more subjective.

Search engines may also flag content because it is similar to published materials as it pulls from the same sources. The AI tools piece content together from various sites and reword them. Without adding the proper flow, this process goes against Google’s “stitching and combining content” guidelines. Content needs to be authoritative and informative, which can be hard to do when piecing information together from various sites without proper human review.

Algorithms devalue content

Google released its helpful content update in August 2022, which highlights “helpful content written by people, for people.” It goes on to state that a search engine crawler looks for content from humans, providing a more cohesive and satisfying practice with SEO.

The update looks to punish content that is created to strictly rank higher in search engine results first. AI tools evaluate SEO results first without truly understanding the text, so the results focus on keywords and not being informative to the reader.

Lack of creativity and personalization

Creative content makes articles more engaging. People tend to share articles they feel a connection to, but AI does not have the emotional intelligence to create a story, instead adding facts to an outline.

AI relies on existing web content and data to develop wording. AI does not understand user intent for queries and still lacks the common sense of human behavior.

Human editing still required

People still need to read through AI-generated content. It might save time, but people still need to be involved and articles quality-checked.

AI tools combine information from several websites into one piece. There may be some mix-ups to fix, such as product descriptions with textures and colors because AI tools do not understand adjective meanings.

Can’t generate new ideas

AI tools use existing data for content, so this means they cannot come up with fresh ideas or original content. AI tools make it hard to come up with new content with the latest, trending ideas and topics.

Spotting AI-Generated Text

Toyama is pessimistic that people can consistently spot the difference between AI-generated and human-generated text, saying, “in the long term, it will become virtually impossible because the AI will become better and better.” He pointed to informal experiments that suggest that even experienced teachers have difficulty distinguishing student writing from ChatGPT.

“I have a colleague who claims that he spotted two instances of ChatGPT submissions because the students’ writing had suddenly dramatically improved—but, ChatGPT can be directed to write at different levels of ability or to introduce errors,” he added.

Computers might be used to identify if a particular text is human-made. Parag Arora, the CEO of Kwegg, an AI content system, told Lifewire in an email interview that research is underway to develop programs that spot machine-generated writing.

“AI classifiers being launched by research organizations, including OpenAI itself, look promising,” he added. “However, GPT advances seem to be always a step ahead of them, and by nature of research, classifiers are always at least one generation behind generators. Some work has been done on leaving cryptic signatures as a policy for every GPT generator, which can be a sure-shot step toward solving this problem.”

There are some ways for users to spot machine-generated text, Robert Brandl, the CEO of Tooltester, said in an email interview. He suggests trying out AI tools like ChatGPT to see what kind of content they generate to get more familiar with them. Among the giveaways are formulaic structures where sentences start with “Firstly” and “Secondly,” etc., and a lack of sarcasm or humor in the text.

“As AI develops, however, the tools that are designed to spot AI text will learn too,” he added. “Indeed, it may get to a point where people need to rely on these tools to understand when AI is being used, that is, if publishers don’t openly disclose AI usage anyway.

Ways to use AI-generated content

AI-generated content is best used as a writing assistant instead of relying strictly on technology. Here are some ways to use AI tools for assistance with content:

Research. For writers having issues organizing a topic or coming up with ideas, AI-generated content can help them get started. Some tools give ideas about what to include for broader topics to help narrow down the research process.
Overcome writer’s block. For writers who know their keyword or topic, AI-generated content tools can help them get started by offering a few hundred words on the subject. Some tools recommend headers so writers can get moving and adapt their content.
Proofread current material. To make sure a drafted article is optimized, writers can run it through AI tools for a grade. The tool can also highlight keywords and phrases that should be used. AI tools can also assist with checking grammar and correcting spelling mistakes.
Write short content. AI tools can produce a lot of content in a short amount of time, so they are a great way to reduce boredom with repetitive tasks. While some communications require more of an emotional side, some short descriptions do not. Product descriptions, metatags, ad copy and social media posts are examples of short text for content generators.
Translate language. For written material to appeal to all audiences, AI generators can help translate content into different languages.
Create templates. AI tools can help create emails or other templates. Some AI tools offer different types of ready-made templates for people to plug in customized information.

Example of AI-generated content

There are many AI content generators available. Here is an example of how it works using a program called CopyAI:

Select style of writing from the left side.
Answer topic.
Add main points.
Select tone.

A couple options pop up for text. Here is the AI-generated content from the information entered. This was the first option meant to imitate the information in this article:

AI is an amazing technology that has the potential to change our world and how we live in it. But like any new technology, there are pros and cons to AI-generated content—and businesses need to be aware of both before deciding whether or not to use it.

First, the pros:

1) AI-generated content saves time and money for businesses by allowing them to produce more content with fewer resources.

2) It allows businesses to create highly personalized content for each customer, which leads to happier customers.

3) AI-generated content makes sure that all information is correct, which helps prevent expensive lawsuits or bad press if something goes wrong with your business’s operations (e.g., if you send out an email with incorrect information).

Now, let’s talk about some of the cons:

1) AI-generated content can sound unnatural (like it was written by a robot), which can make customers feel uncomfortable or distrustful of your business. It doesn’t matter how much effort you put into making sure that everything is correct if the end result sounds robotic!

2) If someone finds out that they’ve been reading an AI-generated piece of content instead.

Much of the information included in this TechTarget article is outlined in the AI-generated content above. However, the AI content is not particularly comprehensive or conversational — even when selecting a friendly tone. This TechTarget article goes into greater detail for explanations and provides examples. The AI-generated content seems to be more of an outline and needs additional information to flow as a human-written article.

There is a larger concern with AI-generated content – students using this tool to cheat. Several professors and teachers stated that the AI-generated content on ChatGPT created convincing essays.

To help distinguish between human and AI text, OpenAI announced a new AI text classifier on Jan. 31, 2023. There are other online tools to help detect AI-generated content by classifying how likely it was written by a person versus AI, such as Originality.ai, Writing.com or Copyleaks.

OpenAI is also working on creating a watermark for longer AI-generated text for an immediate identifier.

In addition to running text on an online AI-content detector, there are a few other signs that content is AI-generated, such as:

Lack of typos. Human writing often has typos and contains slang terms. AI-generated content rarely has typos such as grammar mistakes or misspellings.
Overuse of “the.” AI-generated content is based on predictive language by determining which word should be next, so it commonly uses words such as “the,” “it” or “is” instead of more unusual terms.
No cited sources. AI-generated text will often give facts and not cite the sources.
Shorter sentences. AI content sentence length is typically shorter because it is trying to mimic human writing but appears more choppy than human writing.
Repetition of words or phrases. If certain words or phrases are used abundantly, this could be a sign of AI-generated content because it is trying to fill space with relevant keywords.
Lack of analysis. AI can collect data but cannot analyze the data. If an article feels full of facts without any insight, it might be AI-generated.