Artificial Intelligence

Understanding the Buzz Around ChatGPT and Generative AI

Dr. Muddu Sudhakar

We speak with Dr. Muddu Sudhakar, CEO of Aisera, to understand the hype around ChatGPT and generative artificial intelligence (AI) and their potentials. Sudhakar also talks about what investors should know about these new tools and when he expects wider adoption to happen. 

How is ChatGPT disrupting different industries?  

OpenAI launched this application in early November. So it is too early to forecast a disruption. For example, when Napster and MP3.com came on the scene in the late 1990s, it still took a few years for there to be an impact on the recording industry. 

Although with ChatGPT, the disruption could be accelerated. The reason is that the technology is incredibly powerful and the pace of innovation is breathtaking.  Many people have already had magical moments with how it creates compelling content.    

But we are still in the early innings. The next version of ChatGPT AI is expected to be more powerful by multiples – and it could come out this year. In the meantime, companies like Google and Meta will launch their own large-language models. There are also various startups that are building their own platforms.  

The ChatGPT API will also be essential for disruption. This allows anyone to create their own AI applications or integrate the technology in their existing apps. It has come up with a scripting language for this. OpenAI also has set forth very low pricing for its API, which will help to unleash innovation.    

As for Aisera, we have actually been developing generative AI technology since 2017. We have built our own models that meet the requirements of enterprises, such as for security, compliance, privacy and accuracy. We also provide guarantees for availability. For the most part, ChatGPT remains immature and unstable. Then there are the inaccuracies. Such issues are unacceptable for enterprises.   

What is its potential?  

Generative AI as a category will ultimately revolutionize both consumer and enterprise applications. While much has been said about the consumer applications, enterprises have an opportunity to enable customers and employees to self-serve through conversational dialogues with the expectation to auto resolving those inquiries through AI workflows. 

Many corporate tasks are repetitive and manual; automating them allows employees to focus on more important matters, ultimately leading to increased productivity for the organization. A ChatGPT approach could be a game-changer, as automating a process can be as simple as a prompt like “Turn this knowledge article into a FAQ.”   

How valid is the buzz around ChatGPT?  

ChatGPT is generating a lot of buzz for its powerful ability to generate human-like responses to seemingly any question. The real need from enterprises is a system capable of delivering trust through a high degree of accuracy, context relevance and bound in scope by design. Today, ChatGPT poses limitations in accuracy, lacks context and has broad scope; therefore, it may not always provide correct information, thus undermining trust crucial in an enterprise context.  

If users know ChatGPT’s limitations and use it appropriately, it can be a valuable tool for generating ideas. However, if a user expects reliable and factual information, it may not meet their expectations in its current form.

What is generative AI, and just how big can it get?   

Generative AI uses sophisticated neural networks – which mimic the brain – to create new data, whether text, images, audio or video.  This technology is fairly new. In 2017, a group of researchers from Google published a paper that set forth the architecture for this, which is called the transformer. It’s what powers ChatGPT and most other large-language models (LLMs).  

The transformer is very complex. But on a high-level, it is pretrained on huge amounts of information, such as Wikipedia and Reddit. A major benefit is that there is no need to label this data. The transformer will instead find the relationships, which is done with deep learning systems. From this, the LLM can predict the next word, just like the autocomplete feature on your phone. But since this is based on sophisticated probability theory and sampling, the content is often different, even if the prompt is the same. 

As for how big this can get, it is likely to impact just about every industry. Sequoia Capital predicts that generative AI could ultimately make knowledge workers at least 10% more efficient and creative.    

How might ChatGPT and generative AI impact investors? What should they know?  

Investors should brace themselves for the shake-up that ChatGPT and generative AI technologies are poised to create. With the ability to generate new and original data, these technologies can replace traditional AI/ML models, thereby completely transforming the AI landscape.

Therefore, as an investor, it is crucial to recognize the potential impact of these technologies on the market and keep a close eye on those companies integrating them into their processes. The expectation is that companies that quickly adapt and harness these innovations to create more efficient products and services will enjoy a surge of growth and revenue.

On the flip side, those that are slow to adapt could risk losing market share and being disrupted but more agile competitors.   

When will widespread adoption happen?    

Only two months after its initial release, ChatGPT had already integrated into the software behemoth Microsoft, followed by other powerhouse companies such as Instacart, Shopify and Snap. Their recent announcement to publish APIs certainly accelerate their reach. With this in mind, elements of this technology are not far from widespread adoption as enterprises realize the potential of generative AI for their business needs, provided its boundaries are understood.   

This interview originally appeared in our TradeTalks newsletter. Sign up here to access exclusive market analysis by a new industry expert each week. We also spotlight must-see TradeTalks videos from the past week.

The views and opinions expressed herein are the views and opinions of the author and do not necessarily reflect those of Nasdaq, Inc.

Investing, stocks -- man checking a graph with charts and calculator stock.adobe.com

TradeTalks Newsletter

Sign up to receive your weekly dose of trading news, trends and education

Subscribe

TradeTalks

Nasdaq

TradeTalks is Nasdaq’s regular series covering trading news, market trends and education.

Read TradeTalks' Bio