Large language models are already part of our lives although we are not aware of it

Ever since chatGPT is released, the world can't stop talking about generative AI giving people an excuse to ask if the technology is hyped. ChatGPT, a GPT-based large language model, is all over the news for its benefits and dangers it possesses. Its advanced artificial intelligence capabilities will take every known industry into its stride obliterating the need for human intervention. This idea is certainly hype but what doesn't constitute hype is that it has already become an inevitable part of business operations with only a few downsides like the pricing and data that it requires. The reason: This machine learning model can handle a wide range of use cases. Assistive writing, summarisation, front-line customer care, you name it, Large Language Models are proving to be more than mere tools. Therefore, the hurdles should matter to corporations unless they are blind to tradeoffs. Ben Miller, the CEO of Fundrise, opines, "Although the Large language model hype will be stratospheric, it will birth the greatest productivity boom in American history since the invention of history." Now, this should sound like hype but it is true to the word. To put it in an objective perspective, in a survey held by Tech Crunch, generative AI was voted as the next bubble by most investors.

What Cynics See Vs What Entrepreneurs See

LLMs are developed by large companies and the fear that they would promote incomplete systems as panaceas is very much justified. LLMs having trained to generate human-like answers at times untrue and unverified will be largely used as authentic services by a large number of people giving way to disastrous consequences, and the spread of misinformation. But the optimists have certain reasons to believe in its potential. Even after breaking the utopian expectations, LLMs will be successful in drastically changing the ways of work. For eg., Jarvis, the AI writing tool has already become a fast-growing marketing tool for business houses. Further, it can influence how the non-generative applications of LLMs work. For eg., text classification and named entity recognition(NER) which is currently being used for tasks like extracting information from large amounts of text, will improve noticeably expanding the range of LLM applications.

Can LLMs Deliver the Promise?

To answer this question in a fair manner, one should address the question of the threat LLMs pose to job loss, a question that occupied the big pie in deliberations around chatGPT. Like any tech advancement, though there will be job losses initially, LLMs will only provide an incentive to business models to reinvent their models and practices. No wonder we take for granted the automated suggestions on search engines and uncanny and automated cross-domain replies a bot generates. We are living with generative AI and therefore the question if LLMs are hyped is rather half-baked unless the wider ethical issues have to be taken into consideration. As critics say, they are text prediction models only capable of churning out relatable content. But looking through the entrepreneurial lens lends the models more importance than one can imagine. In spite of the ambiguity around its potential to replicate human behavior, because it gives only an illusion of abstract reasoning, LLMs will live up to the hype.