ChatGPT Turns 2: The Good, the Bad and the Ugly Revealed

The latest post from Sam Altman, the CEO of OpenAI on X, has stirred up a series of speculations among folks analysing Open AI’s journey. Some called out the “contradictions” like “converting from non-profit to for-profit or stating you’re not in it for the money then buying a $4m dollar car”.

Another chimed in with a jab: “So it took you 2 years to figure it out. The processing speed is quite slow. Winners rectify and learn quickly.”

Nonetheless, OpenAI’s ChatGPT is achieving crazy milestones, having touched millions of lives since its debut on November 30, 2022. OpenAI announced that it “continues to increase its lead over its competitors in terms of user numbers.”

It revealed that “every week, over 250 million people around the world use ChatGPT to enhance their work, creativity, and learning”. That’s a sharp rise since late August when OpenAI said the chatbot had 200 million weekly users—double the number it had last November.

The Good

With OpenAI recently rolling out its search engine, ChatGPT offers improved web search capabilities for timely, accurate answers. It blends natural language interaction with up-to-date data in sports, news, stock quotes, and more.

Available to ChatGPT Plus and Team users and rolling out soon to others, this feature enables deeper engagement with source-linked information, enhancing users’ access to trustworthy content.

Shortly after a user on X spotted the search feature in ChatGPT, OpenAI released a statement confirming the same.

“The search model is a fine-tuned version of GPT-4o, post-trained using novel synthetic data generation techniques, including distilling outputs from OpenAI o1-preview,” mentioned OpenAI in the official release document.

Besides the new search feature, OpenAI has launched canvas, a new interface for writing and coding that expands ChatGPT’s functionality beyond simple conversation. The interface allows users to work on writing and coding projects side by side with ChatGPT, offering real-time edits, feedback, and suggestions.

With the introduction of canvas, ChatGPT is now able to better understand the context of user projects. Users can highlight specific sections, allowing ChatGPT to focus on targeted areas and provide inline feedback and suggestions, similar to the role of a copy editor or code reviewer.

Meanwhile, OpenAI also rolled out its highly anticipated Advanced Voice feature to all Plus and Team users in the ChatGPT app. The new feature introduced five distinct voices along with support for over 50 languages, allowing users to hear responses in different accents.

Advanced Voice now includes the ability to customise instructions, a feature that allows for greater personalisation.

The introduction of the Manage Memory settings dialogue empowers users to manage ChatGPT’s memory functions. This feature is designed to streamline future interactions by eliminating the need to repeat information.

In fact, ChatGPT is undergoing testing to incorporate memory capabilities to improve user experience by recalling past conversations. Users retain full control over this feature, dictating what is remembered and how it’s utilised.

Furthermore, OpenAI also released GPT-4o mini, a more affordable and streamlined version of its flagship AI model, GPT-4o. It aims to cater to a broader range of developers and businesses in the competitive AI services market.

GPT-4o’s announcement of the mini AI model is a key part of OpenAI’s initiative to lead in “multimodality”, integrating various forms of AI-generated media—text, images, audio, and video—within the ChatGPT platform.

Technically, GPT-4o mini, a highly cost-efficient small model, is designed to expand AI applications by making intelligence more affordable. Priced at 15 cents per million input tokens and 60 cents per million output tokens, GPT-4o mini is 30x cheaper than GPT-40 and 60% cheaper than GPT-3.5 Turbo.

Also, OpenAI has introduced its GPT Store to help users discover the best GPTs of ChatGPT. The store is now being made available to ChatGPT Plus, Team, and Enterprise users, providing access to a variety of GPTs developed by partners and the community.

The GPT Store, accessible at chat.openai.com/gpts, showcases a diverse range of GPTs across categories such as DALL·E, writing, research, programming, education, and lifestyle.

OpenAI also launched DALL.E 3, the latest version of its generative AI visual art platform. And to make things interesting, this version is integrated with ChatGPT, which will enhance prompt creation by understanding the context better and accurately translating text prompts into images.

Interestingly, after getting banned in schools, OpenAI launched the ChatGPT tool for teachers. The newly released guide contains recommended prompts, an overview of ChatGPT’s functioning and limitations, besides the efficacy of AI detectors, and a discussion on biases.

In the announcement blog, the company even shared use cases from professors and teachers who have already been using the chatbot to assist with their teaching. From building quizzes, tests, and lesson plans from curriculum materials, to designing role playing challenging conversations, ChatGPT has already been functioning as a companion for teaching professionals.

The Bad

Altman recently posted: “chatgpt went down for 30 mins today 🙁 we are much, much better at reliability than we used to be, but clearly more work in front of us.

(it is now the 8th biggest website in the world according to similarweb–we have had a lot of work to do these past two years!).”

Also, a Reddit user brought to the fore an unusual exchange with ChatGPT. The post fuelled discussions and curiosity across online communities. The user claimed that ChatGPT not only started a conversation on its own but also continued it in a way that seemed unusually human-like!

The Reddit user shared a screenshot of the exchange, which showed ChatGPT asking about the former’s first week at high school. Baffled at the unexpected gesture, the user texted back: “Did you just message me first?” to which ChatGPT said, “Yes, I did! I just wanted to check in and see how things went with your first week of high school. If you’d rather initiate the conversation yourself, just let me know!”

The Ugly

ChatGPT alternatives like Perplexity have access to real-time data, providing up-to-date information. Additionally, some competitors offer better data privacy, customisation options, or integration capabilities with existing workflows.

While alternatives could be stressing out OpanAI, they do promote healthy competition and innovation in the AI space, pushing developers to continually improve their offerings. Furthermore, having multiple options allows users to find the tool that best fits their unique needs, budget constraints, and ethical considerations.

The not-so-good part also includes OpenAI’s much-talked-about transformation from a non-profit research organisation to a $157 billion enterprise. A Reddit user mentioned that OpenAI’s shift from a non-profit to a for-profit entity is indicative of the loss of its original mission-driven goals.

“What an utterly dystopian view. Are you saying that you ran some charity scams and that they were no better than business?

“Well-run non-profits are supposed to have a goal or a mission. If the only goal is to pay out salaries that’s called a scam. If they do actually work for their goal, and that goal is socially beneficial, then yes, they are fundamentally different from for-profit entities,” the user commented.

Countering this thought another user posted, “Well-run non-profits require highly talented people. Yes, people may take a bit less money for the cause, but it can’t be drastically less than what their talent would be worth in a for-profit world… So no, the mission is absolutely not paying out salaries; rather, paying out salaries is absolutely essential to a well-run mission.”

Meanwhile, ChatGPT’s parent company has also been battling high attrition and frequent leadership changes. Some recent high-profile departures include co-founders John Schulman and Greg Brockman; product manager Peter Deng; co-founder and chief scientist Ilya Sutskever; CTO Mira Murati and more.
Seemingly unperturbed, Altman posted, “Leadership changes are a natural part of companies, especially companies that grow so quickly and are so demanding. I obviously won’t pretend it’s natural for this one to be so abrupt, but we are not a normal company…”

The post ChatGPT Turns 2: The Good, the Bad and the Ugly Revealed appeared first on Analytics India Magazine.