A blog post on the topic of political discourse

In my opinion, AI Bard just comes up with its own examples even if you ask it to find sentence examples from Internet and newspapers. I am not sure about its quality and reliability in creating conversations. Can you please check whether its English sounds natural and colloquial for you based on some examples.



"Sure, here are some examples of the phrase "Your argument is lacking in nuance" from English corpora:
Example 1:

• Source: A blog post on the topic of political discourse
• Quote: "Your argument that the left is all about identity politics is lacking in nuance. The left is a complex and diverse movement that encompasses a wide range of views on a variety of issues."

Example 2:

• Source: A news article on the topic of climate change
• Quote: "Your argument that the only way to address climate change is to abolish capitalism is lacking in nuance. There are a variety of approaches that can be taken to address climate change, and a one-size-fits-all approach is not likely to be successful."

Example 3:

• Source: A Reddit post on the topic of social justice
• Quote: "Your argument that all police officers are racist is lacking in nuance. There are many good police officers who are committed to serving their communities. We need to address the issue of police brutality without painting all police officers with the same brush."

Example 4:

• Source: A Quora post on the topic of artificial intelligence
• Quote: "Your argument that artificial intelligence poses an existential threat to humanity is lacking in nuance. While AI does pose some risks, it also has the potential to solve many of the world's most pressing problems."

Example 5:

• Source: A Facebook post on the topic of education reform
• Quote: "Your argument that standardized testing is the only way to measure student learning is lacking in nuance. There are a variety of ways to assess student learning, and standardized tests are not the only way to do so.""

I'm not sure about Bard, but ChatGPT is notorious for manufacturing its own examples.

There was even an instance where some lawyer in the US used ChatGPT to research case histories, then cited those cases in his court case. The problem was, he didn't bother to check if any of the cases ChatGPT cited were real, and it turns out none of them actually existed. They were just figments of the AI's imagination. There have been other similar instances of AI manufacturing supposed real-life examples, although that's the one that stuck with me.

Have you tried reverse searching the results? I didn't find any matching results for the first two. Can't you also ask it to cite its sources, then double-check those?

All your examples sound natural enough, but they're too perfectly matched to exactly what you asked for to be real, in my opinion.

Skrej said:

I'm not sure about Bard, but ChatGPT is notorious for manufacturing its own examples.

There was even an instance where some lawyer in the US used ChatGPT to research case histories, then cited those cases in his court case. The problem was, he didn't bother to check if any of the cases ChatGPT cited were real, and it turns out none of them actually existed. They were just figments of the AI's imagination. There have been other similar instances of AI manufacturing supposed real-life examples, although that's the one that stuck with me.

Have you tried reverse searching the results? I didn't find any matching results for the first two. Can't you also ask it to cite its sources, then double-check those?

All your examples sound natural enough, but they're too perfectly matched to exactly what you asked for to be real, in my opinion.

Click to expand...

I asked ChatGPT
How is it to say in up two 4 sentences that your interviews went well using colloquial expressions?

1. "So, my interviews were a breeze! I totally aced them—smooth sailing from start to finish."
2. "Guess what? The interviews went like a charm! I nailed every question and had a good vibe with the interviewers."
3. "I'm stoked! The interviews went super well; it was like we were all on the same wavelength."
4. "Well, the interviews? Piece of cake! I felt in the zone and left with a good feeling about the whole thing."

Would you use them naturally?

They're typical of AI - in a nutshell, overdone. It's trying to cram too many of what are otherwise perfectly natural, casual expressions into one sentence. It almost sounds more like dramatic theater. While I might well use some of the expressions myself, I wouldn't use them all at the same time.

1. "So, my interviews were a breeze! I totally aced them (Stop here). —smooth sailing from start to finish." (adding this sounds trite and superfluous)
2. "Guess what? The interviews went like a charm! I nailed every question and had a good vibe with the interviewers." marginally acceptable
3. "I'm stoked! The interviews went super well. (stop here).it was like we were all on the same wavelength. Unless there were multiple interviewers, use 'both' instead of 'all'"
4. "Well, the interviews? (not at all natural. Use 'Oh' instead of 'Well', if you must) Piece of cake! (Stop here) I felt in the zone and left with a good feeling about the whole thing." unnatural and superfluous. It's unnatural because we say 'we're in the zone', not 'felt in the zone'.

What is it you're after with these? If you're asking whether to trust AI for grammar and naturalness, don't. It's one of the ways as an instructor grading essays I can tell it's at least partially written by AI.

Here's an prime example. I had who student in my college remedial composition class who as struggled with English ( a new language for her). She honestly shouldn't have been allowed to take college courses without first developing a stronger grasp of English, but that's another matter.

As such, she resorted to cheating and plagiarism, including use AI written essays. You know how I new it was AI written (aside from the flowery language and perfect grammar from a student who could barely hold a conversation in English)? The assignment was to write an illustrative essay. She turned in a flawless essay on the history of illustration, going back to monks and illuminated manuscripts.... Besides that, it completely ignored the length requirements by about half.

I'm guess her command of English was weak enough that she couldn't clarify the prompt well enough for the AI.

Skrej said:

What is it you're after with these?

Click to expand...

I aim to express the situations from my own daily experience using natural language. I am striving to turn my own life into the language you speak and otherwise use. What is the best algorithm to follow?

Native speakers. Not AI.

svetlana14 said:

I aim to express the situations from my own daily experience using natural language. I am striving to turn my own life into the language you speak and otherwise use. What is the best algorithm to follow?

Click to expand...

Don't "follow an algorithm"! Study, using good textbooks and with the help of this site. Read really good quality English online - news websites from respected sources such as the BBC or Reuters will help. Listen to English-speaking radio (BBC Radio 4, for example). Watch TV and films in English and try not to use subtitles in your own language. If you really need subtitles, use the English ones.