Brilliant Cosmologist Stephen Hawking once told the BBC, “The development of full artificial intelligence could spell the end of the human race.” Another famous Stephen, Stephen King, wrote “amateurs sit and wait for inspiration while the rest of us get up and go to work.” I would add to that sentiment that people who lack insight turn to A1.
But I would by lying if I didn’t acknowledge that slight pang of dread when a client sent me a description of their company targeted at a specific group generated by Chat GRP-4 and it wasn’t terribly bad. It wasn’t good either. It lacked depth and above all, it lacked the two things only human writers can emote–heart and soul. But it did stitch together bits and pieces it could pull from the web into a plausible accounting.
I decided to ask AI point blank.
So rather than wring my hands and assume the inevitable, I decided to ask AI itself: “Can AI become better at writing than humans?” Its answer might come as a surprise to some who are touting the tech. Here’s what “it” had to say:
“Artificial intelligence (AI) has made significant progress in a variety of language-related tasks, such as machine translation and summarization. However, it is currently not able to write at the same level as a human.
One reason for this is that writing requires a deep understanding of language and the ability to convey complex ideas in a clear and effective way. This requires a level of creativity and nuance that AI systems have not yet achieved.
AI systems can also struggle with context and understanding the intended audience, which are important factors in writing. For example, an AI system may not be able to understand the difference between writing for a general audience versus a technical audience, or the nuances of tone and style that are important in different types of writing.
That being said, AI systems are constantly improving, and it is possible that they may eventually be able to write at a level that is similar to humans in certain contexts. However, it is unlikely that AI will be able to fully replace the creativity and depth of understanding that humans bring to writing.”
Well, shut the front door. While it’s encouraging to say the least, it’s concerning and here’s why.
In that answer, AI hesitates. The line “this requires a level of creativity and nuance that AI systems have not yet achieved. “Yet” is the elephant in the room. AI is supposed to lack self-awareness, and yet somehow it already knows what it is lacking.
So I changed I my line of questioning and asked not whether AI will write better than humans, but rather, what can humans do that AI cannot? I asked, “Can AI be humorous?” And it answered:
“It is possible for artificial intelligence (AI) systems to generate jokes or humorous content, but whether or not the jokes are actually funny is a different matter.
AI systems can be trained on large datasets of jokes and other humorous content, and can then generate new jokes based on patterns and structures it has learned from the training data. However, the jokes that are generated may not always be successful in terms of generating laughter or amusement, because humor is a complex and subjective phenomenon that depends on a variety of factors such as context, culture, and the individual tastes of the audience.
In addition, humor often relies on wordplay, puns, and other linguistic devices that can be difficult for AI systems to understand and replicate. As a result, it is unlikely that AI systems will be able to consistently generate humor at the same level as humans in the near future.”
Whew, so we’ve got that going for us. AI sucks at being funny.
But there’s a much bigger issue than my fragile ego and need for reassurance of my very professional existence. While AI can present challenges and risks for writers, including reducing the authenticity and originality of writing, it also creates major plagiarism issues while it limits creativity and critical thinking. That’s a double whammy of doom for any content someone expects to add credibility to a company. AI can only generate content from content that currently exists somewhere on the Internet. It can also pull content that has questionable ethical and social concerns without any awareness of it. This can affect the accountability and responsibility of writers by blurring the boundaries between human and machine authorship, ownership, and agency.
Because AI relies on data and algorithms for content, the intended tone can easily get lost. AI tools can cover black and white areas of a topic, but gray areas are more subjective. Even more important, AI can’t generate new ideas. That means it can’t come up with fresh thinking or original content, like this article you’re reading right now.
Search engines may also flag AI-generated 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.
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.
Plus, it’s a commonly known fact that creative content makes articles more engaging.
People share content they feel connected to and AI does not yet (there’s that word again) have the emotional intelligence to create a story. Rather, it can only stick together facts from an outline. AI does not understand user intent for search queries and still lacks the common sense of human behavior.
Even if you were to use AI for writing content (perish the thought), you still need a human to review and edit it. Fact-and quality-checking articles still require a human brain and touch. Better yet, start fresh with what you want to cover and leave AI to the duties of research, proofreading, and translation–that’s where it has the greatest potential. Where it goes from there that we have yet to see.