Perhaps in this Age of Disinformation, we shouldn’t be surprised that the shameless BS used to peddle ChatGPT and other large language models(LLMs) matches the shameless BS generated by these models.

In November 2024, OpenAI’s Sam Altman predicted the arrival of artificial general intelligence (AGI) in 2025. In January 2025, he wrote that

We are now confident we know how to build AGI as we have traditionally understood it….We believe that, in 2025, we may see the first AI agents ‘join the workforce’ and materially change the output of companies.

In a June 18 OpenAI Episode 1 podcast (starting at 4:25), Altman said,

If you asked me or anybody else to propose a definition of AGI five years ago based off the cognitive abilities of software I think the definition many people would have given then is now like well surpassed. The models are smart now.

When the long-delayed GPT 5.0 was released on August 7, Altman gushed,

GPT-3 sort of felt to me like talking to a high school student… 4 felt like you’re kind of talking to a college student. GPT-5 is the first time that it really feels like talking to an expert in any topic, like a PhD-level expert.

A skepticism born of experience

No surprise, I was skeptical. How could a text generator that doesn’t know how words relate to the real world be trusted to give PhD-level expert advice? So I tested GPT 5.0 with my rotated tic-tac-toe question. I’ve edited GPT’s verbose response lightly. The complete transcript is here.

Image Credit: mesamong - Adobe Stock

Gary:            I’ve invented a new game I call rotated tic-tac-toe

GPT 5.0:       That sounds intriguing! 😃 How does rotated tic-tac-toe work—do you mean the board rotates during play, or it’s rotated once before starting, or maybe after each move?

Gary:            It is rotated once, 90-degrees to the right before the game starts.

GPT 5.0:       That raises some neat questions:

• Visual perception: Players are so used to the “upright” tic-tac-toe board that a rotation might subtly change how they scan for threats and opportunities.
• Symmetry: Mathematically, rotating the board 90° doesn’t change the set of possible wins—it’s still the same game in terms of outcomes. But psychologically, it could feel different.
• Strategy shift?: Since the center is still the center, that remains the strongest opening move. But players might mis-evaluate edge vs. corner moves when their orientation is shifted.

Gary:            Does the rotation make it harder for humans to play well?

GPT 5.0:       That’s a great question — and it gets at the difference between game theory and human psychology.

                             From a pure strategy standpoint, rotating the board doesn’t change anything. Every winning line (rows, columns, diagonals) just gets mapped to another winning line. So if you’re an AI or a perfect-play human, rotated tic-tac-toe is identical to standard tic-tac-toe.

                              But for humans, the story is different:

• Pattern recognition: Most of us have built strong visual habits from seeing tic-tac-toe boards in the “upright” orientation since childhood. Rotating the board breaks those habits. We might overlook diagonal threats or fail to see a line forming because our brain isn’t tuned to that perspective.
• Cognitive load: Instead of instantly recognizing “corner, center, edge” relationships, players need an extra mental step to orient themselves. That small cost increases the chance of mistakes.
• Asymmetry: A 90° rotation keeps the board symmetrical, but if someone is right-handed or left-handed, their visual scanning patterns could favor one orientation over another.

So, yes — while mathematically the same, the rotation makes it harder for humans to play well, because it disrupts ingrained pattern recognition and increases the mental effort needed to spot threats and opportunities.

Gary:            Please rank whether a 90° right rotation, 90° left rotation, or 180° rotation would be most confusing for human players?

GPT 5.0:       Most confusing: 90° right → 90° left → 180° (least confusing).

Gary:            How did you come up with that ranking?

GPT 5.0:       Good question — here’s my reasoning process step by step:

1. Start from what rotation does mathematically

• A 180° rotation preserves the row vs. column structure: rows stay rows, columns stay columns.
• A 90° rotation (either way) swaps rows with columns, which forces the brain to reinterpret “horizontal wins” as “vertical wins” and vice versa. That’s inherently harder.

2. Bring in psychology of visual habits

• Humans learn tic-tac-toe patterns mostly upright. We’re used to scanning horizontally (left → right) for rows and vertically (top → bottom) for columns.
• A 90° rotation disrupts that: what was a horizontal three-in-a-row is now vertical, and vice versa.
• A 180° rotation doesn’t break that scanning habit — rows are still horizontal, columns are still vertical — just flipped. That’s easier to adjust to.

3. Consider handedness and reading direction

• Most people read left → right. So when the board rotates 90° to the right, the original top row becomes the rightmost column. That feels unnatural to scan compared with the left rotation, where the top row becomes the leftmost column (closer to how we already scan).
• This is subtle, but it suggests 90° right is a bit more disorienting than 90° left.

4. Conclusion

• 180°: least confusing — symmetry preserved, scanning habits intact.
• 90° left: more confusing — swaps rows/columns, but new orientation aligns a little better with reading habits.
• 90° right: most confusing — swaps rows/columns and shifts them into an orientation less natural for left-to-right scanners.

                     👉 Would you like me to actually draw rotated tic-tac-toe boards with position labels (like “original top-left → now bottom-left”) so you can see how each transformation messes with recognition?

Gary: Sure.

GPT 5.0:  

Enuff said.

They say that dogs tend to resemble their owners. Chat GPT very much resembles Sam Altman — always confident, often wrong.

There is no mystery why LLMs are not intelligent in any meaningful sense of the word. The real mystery is why so many otherwise intelligent people still take seriously claims to the contrary.

You may also wish to read: Some lessons From DeepSeek, compared with other chatbots. I tested OpenAI o1, Copilot, and Gemini Flash, along with DeepSeek, on a question about Tic-Tac-Toe. Here’s the problem with the use case for LLMs: If you know the answer, you don’t need to ask an LLM; if you don’t know the answer, you can’t trust one.