Listen to this Post
Earlier this week, I found myself in a classic parenting moment. My son, full of curiosity, asked me a question that seemed simple enough: “What is the name of the shape that looks like a square with uneven sides?” As I thought about the answer, I realized that the situation could be the perfect opportunity to test out a new feature—ChatGPT through Siri. What followed was a surprising mix of voice recognition errors, unexpected results, and, ultimately, an exploration into how these technologies interact.
The Journey Begins: Siri and ChatGPT Together
It all started during our drive to school, when my son’s question about a shape with uneven sides piqued my interest. For questions like these, I’ve gotten into the habit of telling him, “Ask ChatGPT.” However, on this occasion, instead of using the typical text-based input, I decided to try something new: ChatGPT through Siri.
I picked up my iPhone, pressed the Action button, and launched ChatGPT’s Advanced Voice Mode. The results were immediate, with ChatGPT providing a satisfying answer to my son’s question. But, this was just the beginning. As I delved deeper into the experiment, I found that using Siri to ask ChatGPT for an answer led to some unexpected outcomes, all of which left me with more questions than answers.
The Quirks of Siri and ChatGPT Integration
I began by running the same query through ChatGPT, first using its Advanced Voice Mode. The conversation, while straightforward, had a funny twist: the voice-to-text feature misheard “shape” as “sheep,” yet ChatGPT quickly kept the conversation on track. Despite this minor hiccup, I was impressed with the response.
But then came the experiment with Siri. I used the voice assistant to ask ChatGPT the same question: “Hey Siri, ask ChatGPT what is the name of the shape that looks like a square with uneven sides?” My iPad, in particular, threw a curveball. Instead of answering, Siri asked if I wanted to send content from an open Safari tab to ChatGPT. My iPhone, on the other hand, repeatedly asked if I wanted to send a screenshot of my screen. At first, I was baffled—why was Siri not understanding the question?
It slowly dawned on me that Siri might be interpreting the request as something related to the visual content on my screen, possibly thinking I was asking about a shape displayed there. Clever? Yes. Correct? Not quite. This mix-up made me wonder if I had stumbled upon a new layer of interaction between Apple’s virtual assistant and the AI chatbot.
ChatGPT’s Straightforward Answer
After these trials with Siri, I decided to go back to basics and just ask ChatGPT directly, without any interference from voice assistants. This time, it answered the question with precision: a rhombus seemed like the best fit for the description. However, my search didn’t stop there. A quick online search revealed that “rhomboid” was also a possible answer, but another shape—trapezium—could be a contender depending on the definition.
By now, I was left with a growing uncertainty about the nature of my son’s original question. In the end, I had more confidence in the Google search results, which pointed more decisively toward “rhombus” than anything else. But still, was that the definitive answer? And did Siri really fail, or was this just a misunderstanding of intent?
What Undercode Says:
This exploration into Siri and ChatGPT reveals how different digital assistants can interpret user input in unexpected ways. While Siri is highly integrated into Apple’s ecosystem, it often seems to struggle with more abstract queries—especially when they involve specialized knowledge. The confusion over whether I wanted to send content to ChatGPT or the interpretation of my question as related to on-screen content shows that Siri may not yet have the contextual understanding required to handle complex inquiries effectively.
On the other hand, ChatGPT’s responses were consistently helpful when used directly, offering a deeper level of understanding and flexibility. This indicates that AI chatbots, when allowed to function without the constraints of other voice assistants, can provide more accurate and detailed information. The issue here, however, isn’t about the capability of ChatGPT itself but the nuances of integrating such AI with existing systems like Siri.
In a broader sense, the experiment highlights the importance of clarity in digital communication. While both Siri and ChatGPT are impressive tools, their effectiveness is still largely dependent on how they interpret user queries. Siri’s reliance on the assumption that a user might want to share content from a web browser suggests that voice assistants still have a long way to go in terms of contextual understanding. Meanwhile, ChatGPT’s ability to understand and respond to a variety of queries without context-based limitations shows why AI chatbots might soon become a preferred way to seek answers, especially for open-ended or complex questions.
While the misinterpretation of the shape query may seem minor, it underscores a larger problem within voice-based AI technologies: how do we ensure they understand our intent? This question remains as relevant as ever, especially as we rely more on these tools in our everyday lives.
Fact Checker Results
- The shape discussed is most commonly referred to as a rhombus, but rhomboid and trapezium are also valid answers in specific contexts.
2.
- ChatGPT provided the most accurate response when directly asked, showcasing its strength in handling detailed queries.
References:
Reported By: 9to5mac.com
Extra Source Hub:
https://www.twitter.com
Wikipedia
Undercode AI
Image Source:
Pexels
Undercode AI DI v2
