Accenture
- Generative AI will empower industries beyond Big Tech, according to Accenture's data lead.
- "It is not just on the onus of a couple of really strong tech companies to unlock it," Teresa Tung said.
- Tung is also working with industry bodies to establish responsible AI frameworks.
Big Tech will lose some of its stranglehold on breakthrough technologies thanks to the rise of generative AI, which will empower smaller companies, according to the data lead at consultancy firm Accenture.
Generative AI, the name for a range of tools that can create computer-generated content based on large pools of data, has become enormously popular since the release of ChatGPT late last year. OpenAI's buzzy language-generation tool is trained on 570 GB of data taken from books and the internet.
Its terms of engagement are simple: Ask a question or give it a prompt, and it will reply. The viral bot has been a hit with internet users, with many generating books or simply tormenting it for fun. Generative AI companies have secured over $1.7 billion of venture capital investment over the last three years, according to Gartner. The wider AI market is tipped to reach $134.8 billion by 2025.
Generative AI can be fed with examples of business documents, proposals, and code, amongst other things, to generate fresh assets. Its ease of use and ability to churn out content, which actually considers the context of queries, makes otherwise complex technology like AI accessible to small companies.
"The big thing is, it's democratized," Teresa Tung, a chief cloud technologist at Irish American consultancy firm Accenture, told Insider during an interview at Mobile World Congress 2023. "It is not just on the onus of a couple of really strong tech companies to unlock it."
This allows businesses to grow in a new way; before generative AI, the pace and scale at which companies could implement AI may have been restricted by headcount as specialist employees, such as data scientists, were required.
"You see how easy it is to apply ChatGPT; the application becomes the new pivot in terms of focus," Tung, who has over 220 patents and patent applications in her name, said.
"So rather than having to scale by having lots and lots of data scientists, maybe you have more and more domain experts and people who are getting the data ready for the data engineers."
In healthcare, for example, doctors should be able to pinpoint what data they have and how to apply it. "I don't necessarily need a whole bunch of data scientists to be able to create a model now that is used for predicting patient outcomes," she said. Instead, doctors could do this using generative AI. "That's really exciting. Having the data, [and] having the domain experts becomes more important than having the data scientists."
This pivot gives power back to industries outside of tech, Tung said, as they are the ones with both domain data and expertise. "The challenge is their ability to, one, figure out how to apply this responsibly, and then, two, be able to engineer it within their human processes in big central use cases."
The technologist stressed that there should always be a human in the loop to check if the AI-generated information is correct. Tung is also working with industry bodies to establish responsible AI frameworks.