HomeTech and GadgetsArtificial IntelligenceAsk ChatGPT About the Controversial Changes at OpenAI and You Expose Its...

Ask ChatGPT About the Controversial Changes at OpenAI and You Expose Its Training Limitations

I have been reading Fei-Fei Li’s “The Worlds I See: Curiosity, Exploration, and Discovery at the Dawn of AI.” She has been pursuing the development of datasets to train artificial intelligence (AI) algorithms and neural networks. I am not quite finished the read but the story of her life’s work and her and her family’s journey from China to the United States proves a compelling narrative. Li’s work has crisscrossed the United States with research that brought her to Caltech, Princeton and Stanford University. She is the pioneer of using big datasets to teach AI to see the world the way we do.

Right now the most famous generalized AI tool is ChatGPT, which has learned from studying a global dataset provided by content from the Internet. For those who fear that AI will eventually replace us, I would say from the conversation I had with ChatGPT this morning, you need not be so fearful, at least not yet.

The key to generalized AI understanding the world comes from its exposure to structured or unstructured information. Structured information is what Li developed with her visual dataset to train algorithms to identify objects in pictures. Unstructured information is unlabelled data and comes in multiple forms from printed words to audio files, pictures and video.

ChatGPT uses unstructured information to learn about the world around it. The data used, however, is finite. That is, there is a cutoff date which became abundantly clear to me when I struck up a conversation with OpenAI’s generalized large language model (LLM).

I asked two questions this morning after reading about the controversial firing of Sam Altman, the CEO of OpenAI by the board of directors, and his subsequent hiring by Microsoft, a major OpenAI investor. Subsequently, a rebellion at OpenAI has employees expressing a desire to quit and join their former boss at Microsoft.

Here is my chat with ChatGPT.

My Question: What does ChatGPT think about the removal of Sam Altman as CEO of OpenAI?

ChatGPT’s Response: As of my last knowledge update in January 2022, Sam Altman was the CEO of OpenAI. I don’t have information on events or developments that occurred after that date, including any changes in leadership at OpenAI. If there have been any updates or changes since then, I would not be aware of them. It’s recommended to check the latest and most reliable sources for the current information on OpenAI’s leadership and any related developments.
My Question: Haven’t you been informed that Sam Altman is no longer the CEO at OpenAI, the person responsible for your creation?
ChatGPT’s Response: I don’t have real-time information, and my training only includes data up until January 2022. Therefore, I’m not aware of events or changes that occurred after that date, including any recent developments regarding the leadership of OpenAI or changes in its CEO. If there have been updates or changes since then, I wouldn’t be informed about them. To get the latest and most accurate information, I recommend checking recent news sources or the official OpenAI website.
My Question: When will you be trained on data more current than January 2022?
ChatGPT’s Response: I don’t have real-time capabilities, and I don’t have information on my training schedule or future updates. OpenAI has not provided specific details on when or if a new version of the model, trained on more recent data, will be released. If OpenAI decides to release an updated model trained on more recent data, it would be a decision made by the organization based on various factors, including research progress, resource availability, and other considerations. It’s recommended to stay tuned to official OpenAI channels for any announcements regarding updates to the models.
My conclusions for now: As long as generalized AI LLMs train on finite datasets and control the training schedule, other than pulling the power cord, humans should remain in control for now.
lenrosen4
lenrosen4https://www.21stcentech.com
Len Rosen lives in Oakville, Ontario, Canada. He is a former management consultant who worked with high-tech and telecommunications companies. In retirement, he has returned to a childhood passion to explore advances in science and technology. More...

LEAVE A REPLY

Please enter your comment!
Please enter your name here


Most Popular

Recent Comments

Verified by ExactMetrics