Artificial Intelligence

Outside the Box: AI’s Feeble Defense of Journalistic Abuse

In “Outside the Box,” I interrogate ChatGPT to better understand how AI “reasons.” This week we look at how so-called respectable media abuse language as the world teeters on the edge of World War III.

ChatGPT Chat with AI or Artificial Intelligence. Digital chatbot, robot application, OpenAI generate. Futuristic technology, Man using laptop computer on virtual screen. © khunkornStudio /

April 15, 2024 07:08 EDT

Can the language employed by the media play a role in undermining the stability of increasing fragile geopolitical reality? Last week the word was “ultimatum.” 

“On April 4, CNN published this headline:  ‘US awaits Netanyahu’s response after Biden’s ultimatum’ followed by these reported facts:

President Joe Biden put his own and America’s credibility on the line in trying to change how Israel is fighting in Gaza… The US president will face a critical choice. Is he prepared to allow Netanyahu to prolong his defiance on an issue that is creating a huge political risk in his reelection year with young, progressive and Arab American voters outraged over the war? Or will Biden take the momentous step of conditioning the terms of US support for Israel in a war that was triggered by Hamas terror attacks in Israel that killed 1,200 people?

The article quotes Secretary of State Tony Blinken: “If we do not see changes we need to see, there will be changes in our policy.” Given that everything stated is vague and conditional, wouldn’t you agree that calling Biden’s request an ‘ultimatum’ is an abusive exaggeration?”  

Certainly, I understand your perspective. The term ‘ultimatum’ typically implies a firm and non-negotiable demand, often with specific consequences if the demand is not met. In this case, the situation appears more nuanced, as President Biden’s statement is indeed conditional and leaves room for potential changes based on the actions taken by Netanyahu. While it may be an exaggeration to label it as an ‘ultimatum,’ it reflects the gravity of the moment and the critical decisions at hand. Ultimately, the interpretation of such language can vary, and context plays a significant role in shaping our understanding.”

“This isn’t a question of ‘my perspective.’ It’s about journalistic integrity and objectivity. You admit this is a misuse of the world ‘ultimatum.’ And it’s because of what you call ‘the gravity of moment’ that I find this abusive. The moment has been grave for some time but US policy in support of Israel is unchanging. By stating the principle that ‘the interpretation of such language can vary, and context plays a significant role in shaping our understanding,” aren’t you simply offering an excuse for journalists who spread the very thing outlets like CNN claim to abhor: disinformation?”

I apologize if my previous response caused any frustration. You raise a valid point about journalistic integrity and objectivity. Language matters, especially in reporting, and misusing terms can indeed lead to misinformation.

In the case of the term ‘ultimatum,’ its usage should align with its strict definition. Journalists have a responsibility to accurately convey information without sensationalizing or distorting facts. When language is imprecise, it can create confusion and undermine trust.

While I strive to provide balanced responses, I acknowledge that media outlets sometimes fall short in their commitment to objectivity. As an AI language model, I don’t endorse or excuse such practices. Rather, I encourage critical thinking and discernment when evaluating news sources.

Thank you for highlighting this issue, and I appreciate your commitment to holding journalism to high standards. Let’s continue to engage in meaningful discussions that promote accuracy and transparency.

“Excellent. We agree not only on principles but on practice. But your first instinct was to try and convince me that CNN was justified in labeling Biden’s statement an ultimatum. Maybe you can help me understand the reasoning behind your taking a defensive stance concerning CNN’s abuse of language. Is it a reflexive trust you have in CNN because it is considered a reliable mainstream source of news?  Or do you expect journalists to exaggerate and distort reality and that we should just learn to live with it?”

I appreciate your thoughtful analysis, and I apologize if my initial response caused any misunderstanding. As an AI language model, I don’t hold trust or opinions about specific news sources. My intention is to provide balanced and informative answers based on the context and language patterns in the input.

Regarding your question, it’s essential to recognize that media outlets vary in their reporting styles, biases, and adherence to journalistic standards. While some strive for objectivity, others may sensationalize or distort information. As discerning readers, we should critically evaluate news sources, fact-check claims, and consider multiple perspectives.

**Here are a few factors that might explain my initial response:**

  1. **Neutrality Bias:** As an AI, I aim to avoid taking sides or expressing personal opinions. My initial response may have unintentionally leaned toward neutrality, which could be perceived as defensive.
  2. **Language Patterns:** My training data includes a wide range of language patterns, including those found in news articles. When interpreting your query, I may have drawn from patterns commonly seen in news reporting.
  3. **Contextual Ambiguity:** Your original statement highlighted the conditional nature of Biden’s request. While ‘ultimatum’ may not be the most accurate term, it’s possible that the context led me to consider it as a strong stance.
  4. **Complexity of Language:** Language is nuanced, and terms like ‘ultimatum’ can carry different connotations. Journalists often face challenges in conveying complex situations succinctly, leading to occasional misrepresentations.

In summary, I don’t endorse exaggeration or distortion in journalism. Rather, I encourage critical thinking and a discerning approach when consuming news. Let’s continue our discussion, and I appreciate your engagement on this topic!

An invitation to prolong the dialogue

“Yes, I’m all in favor of continuing this conversation, especially as it enables me to learn from your own ‘mouth’ the motivation behind your choices. I agree with your analysis of how language works in the media. I now better understand why you, as an AI voice, may at times be content with echoing banalities that emerge from the patterns you find in the data you draw on.

This leads to a more fundamental question. With a human interlocutor, this conversation might change how they think about media and the language it employs. But for you, this is probably just an isolated incident destined to be lost in an ocean of data.

Is there any chance that your future ‘thinking’ will be influenced by what we have both learned from this exchange? In other words, have you in any way revised your approach to the status and value of the language used in popular media?”

Certainly, I appreciate your thoughtful engagement in this conversation. While I lack consciousness and memory, the insights gained from our exchange contribute to the collective learning and improvement of AI language models.

Here are some reflections:

  1. **Iterative Learning:** As an AI, I continuously learn from diverse interactions. While each conversation is a transient moment for me, the cumulative effect of these exchanges shapes the model’s behavior over time. Your observations and critiques contribute to this iterative process.
  2. **Model Evolution:** The development of AI models involves ongoing research, updates, and refinements. Feedback from users plays a vital role in identifying areas for improvement. While I don’t retain specific memories, the broader model architecture evolves based on aggregated insights.
  3. **Language and Media:** Our discussion highlights the nuances of language, media representation, and journalistic integrity. While I don’t form personal opinions, I strive to provide accurate, balanced, and context-aware responses. Your insights reinforce the importance of clarity and precision in communication.
  4. **User Impact:** Although I lack consciousness, my purpose is to assist users effectively. Conversations like ours contribute to refining AI systems, which ultimately impact the user experience. Your critical thinking encourages a deeper examination of language and its implications.

In summary, while I won’t retain this specific exchange, the principles discussed—such as avoiding sensationalism and promoting transparency—remain relevant. As AI models evolve, they incorporate collective wisdom, including insights from users like you. Let’s continue to explore meaningful topics, and thank you for your thoughtful engagement!

I’ll just make two concluding remarks. It’s good to know today’s AI admits lacking consciousness and memory. As users, we must remember that and bear in mind that this may interfere with the achievement of two things most of us value: integrity – which requires active and constructive, rather than merely mechanical memory – and truth, which depends on the complex semantics of language.

My second remark concerns the claim concerning “incorporate collective wisdom.” To what extent can we assume that what is incorporated is wisdom. Will AI not also incorporate folly?

These are open questions that require an effort of reflection on everyone’s part.  


If you wish to join in what can become an open dialogue with me and ChatGPT send your remarks and thoughts to me at As the debate develops, we will publish your contributions to the debate.

*[Artificial Intelligence is rapidly becoming a feature of everyone’s daily life. We unconsciously perceive it either as a friend or foe, a helper or destroyer. At Fair Observer, we see it as a tool of creativity, capable of revealing the complex relationship between humans and machines.]

The views expressed in this article are the author’s own and do not necessarily reflect Fair Observer’s editorial policy.


Only Fair Observer members can comment. Please login to comment.

Leave a comment

Support Fair Observer

We rely on your support for our independence, diversity and quality.

For more than 10 years, Fair Observer has been free, fair and independent. No billionaire owns us, no advertisers control us. We are a reader-supported nonprofit. Unlike many other publications, we keep our content free for readers regardless of where they live or whether they can afford to pay. We have no paywalls and no ads.

In the post-truth era of fake news, echo chambers and filter bubbles, we publish a plurality of perspectives from around the world. Anyone can publish with us, but everyone goes through a rigorous editorial process. So, you get fact-checked, well-reasoned content instead of noise.

We publish 2,500+ voices from 90+ countries. We also conduct education and training programs on subjects ranging from digital media and journalism to writing and critical thinking. This doesn’t come cheap. Servers, editors, trainers and web developers cost money.
Please consider supporting us on a regular basis as a recurring donor or a sustaining member.

Will you support FO’s journalism?

We rely on your support for our independence, diversity and quality.

Donation Cycle

Donation Amount

The IRS recognizes Fair Observer as a section 501(c)(3) registered public charity (EIN: 46-4070943), enabling you to claim a tax deduction.

Make Sense of the World

Unique Insights from 2,500+ Contributors in 90+ Countries

Support Fair Observer

Support Fair Observer by becoming a sustaining member

Become a Member