Can You Pass the "Kitchen Table" Test? 💜

The “Kitchen Table” Test! 😊

#newspaperinlearning

The New Indian Express

24 April 2026

This morning I was reading this particular article in today’s The New Indian Express!

It’s about Anthropic’s “Claude Mythos” model and hence I was curious to know more about it!

On an aside, as most of us regularly following the English dailies this past one month would have known, Claude Mythos has been making it to the news for all the wrong reasons!

For some background information on the subject for us all –

Well, unlike our regular AI models like Gemini or ChatGPT or even the previous versions of Claude, Claude Mythos has not been released to the general public. The reason is that, the model has unprecedented, highly advanced cybersecurity capabilities, and hence it has the potential to autonomously write code, and completely hijack systems of literally anybody in any part of the world!

Hence, Anthropic gave the world at large a subtle warning that, releasing Claude Mythos publicly would be too dangerous, as it would literally give anyone in the world the power to launch cyberattacks. Instead, they said they are restricting it only for a closed programme called “Project Glasswing,” allowing only select partners (like tech giants and banks) to use it to patch their systems before hackers can hack into them.

However, as irony or ‘unluck’ would have it, there was a huge security lapse that had happened here.

Anthropic’s heavily guarded cybersecurity model was leaked by some unauthorised hackers - users who managed to gain access to the restricted Mythos preview! Anthropic has also confirmed the breach. This has caused great panic among the cybersecurity community, raising serious questions about whether AI companies are actually capable of keeping high-risk AI Models from the hands of the illegal, dangerous hackers!

It is in this context that I wanted to read this article with all curiosity!

However, one cursory look at the article, and I could feel that, the writing is highly unreadable! ☹

This kind of writing style is usually dubbed in academic circles, as the result of a “curse of knowledge,” where the author assumes that the reader already possesses a deep understanding of all the complex concepts in cybersecurity that he has ‘discussed’ in this article.

So what exactly is wrong with the writing style here?

The author – an engineering graduate - has used a lot of highly technical terms without taking time to explain them in plain English, or providing any contextual clues or definitions whatsoever!

Sample some of the phrases like “autonomous exploit chaining”, “semantic mapping”, and “privilege escalation paths” – terms that read like they are meant for security engineers, and NOT for a news article that would cater to a general audience!

A layperson reading this article might become thoroughly confused and perplexed by all means! Most of the points in this article are likely to fly completely over their heads! 😊

Secondly, the article does not paint a visual picture for the lay reader, which makes it sound so abstract and twice removed from the lay!

Sample this: “A model can sustain analysis across large software environments without losing continuity” or “The first ramification is visible in reconnaissance, and this creates a distinct threat vector for cyber security”.

These sentences sound quite scholastic, bombastic and impressive! 😊 However, they lack the readerly-connect in that, they do not aim at painting a visual picture for the lay person!

As such, the article fails miserably in the proverbial “kitchen table test”.

So what pray, is the Kitchen-Table Test?

The “kitchen table test” has been a universal benchmark of sorts, in academic writing, journalism, and in politics to check whether a concept, or a news story is relatable and easily understood by the average person!

To pass the test, the subject matter must be quite simple in style – so that one could easily explain it to a friend or family member during a casual conversation over dinner at the kitchen table, without relying on the highly jargonised vocabulary! It simply means that, if ordinary people wouldn’t talk about it at their kitchen table, it definitely needs to be rewritten!

As the legendary Albert Einstein so beautifully puts it -

If you can’t explain it simply, you don’t understand it well enough! 😊

So yes, beware of the “kitchen table test” when you speak or write! 😊

PS: You may want to read a humorous take by one of my favourite writers – P. Sainath - on one such jargonised vocabulary used by NGOs and the Insurance / Banking / Health Sectors, HERE on our blog, written well over 15 years ago!