Virtualization Technology News and Information
Article
RSS
Deepgram 2023 Predictions: Trends in innovation-driven technology, generative AI and data-driven solutions

vmblog-predictions-2023 

Industry executives and experts share their predictions for 2023.  Read them in this 15th annual VMblog.com series exclusive.

Trends in innovation-driven technology, generative AI and data-driven solutions

By Scott Stephenson, CEO and co-founder of Deepgram

In the current financial downturn, many tech companies are finding themselves at a crossroads between cutting costs and investing in innovation. The reality, however, is that they don't have to choose one strategy over the other. Although some parts of the economy are indeed cooling off, demand for technology solutions to business problems will remain strong in 2023. 

As the CEO and co-founder of Deepgram, an industry-leading speech technology company, I've got a front seat to the technical trends and innovators that are reshaping the business landscape with artificial intelligence. The following are a few of my predictions for 2023, including why AI is recession proof, key tech trends to follow in the new year and why I think that end-to-end deep learning needs data-driven solutions. 

AI is Recession Proof

Despite the economic uncertainties that consumers and companies are facing, I predict that demand for AI-enabled products and services will remain not only resilient but recession-proof. 

Over the past five or so years, business leaders enjoyed a permissive financial environment. "Growth at all costs" was the mandate, made affordable by all the cheap, plentiful capital flying around the market at the time. Well, we all know what happened next: inflation picked up and the Federal Reserve took away the punchbowl. Party over, for now.

The U.S. economy is not in a recession, yet, but business leaders must deal with a new set of financial constraints while still managing to eke out some business growth while they still can. Since a company can no longer spend its way out of a business problem, innovation and automation is now the name of the game. 

Simply put, AI will help us do more with less. Artificial intelligence in the workplace has come a long way from the little animated paper clip in the corner of your word processor. AI solutions save teams hundreds or even thousands of hours by automating tedious processes, and they enable companies to access and use their data in new ways that weren't feasible without large-scale computation and predictive modeling. Working with AI is not the far-out science fiction future; it's happening at basically every large company today. I predict that AI will continue to be central to business in 2023 by cutting costs and enabling innovation.

All signs point to continued deterioration of macroeconomic conditions as the effects of mid-to-late 2022's interest rate hikes start to come online, so 2023 may bring a serious test of this hypothesis.

The Next AI Wave is Generative 

In 2023, one of the biggest focuses will be on generative AI technologies. This really does feel like something new, and the rate of progress is nothing short of amazing.

At this point you've almost certainly heard about AI programs-like OpenAI's DALL-E 2 model, Google's Imagen model, or the open source image synthesis toolkit Stable Diffusion-that take a written description of an image and turn it into an actual image, with some models also supporting video generation. By that same token, you might've had a conversation with ChatGPT, a chatbot from OpenAI that relies on a large language model to produce eerily human-sounding responses to basically whatever question or challenge you throw at it. 

It's kind of hard to believe that these tools were only released in 2022. It's taken years of fundamental research and development work to get to this point, and if 2022 was the year that a lot of folks started talking about generative AI, I predict that 2023 will be the year when even more folks start using generative AI.

Although the technology seems relatively siloed off in the world of art or content creation, for now, generative AI has applications throughout the economy-from banking and industrial design to data security and personalized marketing, and beyond! It will be exciting to see which new use cases are discovered in the next year, and I wouldn't be at all surprised if generative AI plays a role in making some of those discoveries.

End-to-End Deep Learning Needs Data-Driven Solutions

2023 will be the year of end-to-end deep learning in the enterprise. No longer a speculative area of academic computer science research, the stuff of skunkworks projects that will never see the light of day, or a mere component in a complex data processing pipeline, most companies today have enough data to build powerful models that can handle all the steps of a given business process. The pace of progress in applied deep learning is accelerating, so what may be on the cusp of possibility today will be table stakes in a couple of years. Deep learning models have proven to be a remarkably robust part of the new business stack. 

Overall, it is clear that technology is resilient despite economic circumstances, and it is crucial that technology leaders can identify where change and creativity can happen in order to rise above these challenges. In 2023, AI will continue to help companies drive growth, innovatively disrupt the way we work, and adapt to the ebbs and flows of the way humans interact and work. 

##

ABOUT THE AUTHOR

Scott Stephenson 

Scott Stephenson is a dark matter physicist turned Deep Learning entrepreneur. He earned a PhD in particle physics from University of Michigan where his research involved building a lab two miles underground to detect dark matter. Scott left his physics post-doc research position to found Deepgram.

Published Wednesday, January 18, 2023 7:35 AM by David Marshall
Comments
There are no comments for this post.
To post a comment, you must be a registered user. Registration is free and easy! Sign up now!
Calendar
<January 2023>
SuMoTuWeThFrSa
25262728293031
1234567
891011121314
15161718192021
22232425262728
2930311234