Innovation and Creation in Ever-Advancing Artificial Intelligence

By Ray Schroeder

We understand that artificial intelligence (AI) resulted in the loss of many blue-collar jobs as smart robots took over the manufacturing process. However, we now know that this generation of AI will have even greater impact in truly creative fields, including art and original authorship. The impact is far-reaching and revolutionary.

Decades ago, it was broadly predicted that digital machines would sweep the workforce world, leaving precious few, mostly white-collar-worker, positions. The professionals, managers, and executives would continue to work in their higher-paid, benefits-laden positions; the rest of the workforce would be replaced by AI. At first, in large part, the advent of AI did impact the non-professional positions at a higher rate than the skilled professions. Economist and public policy analyst Professor Harry J. Holzer of Georgetown University writes, “Indeed, digital automation since the 1980s has added to labour market inequality, as many production and clerical workers saw their jobs disappear or their wages decline. New jobs have been created – including some that pay well for highly educated analytical workers. Others pay much lower wages, such as those in the personal services sector.”³

In the intervening years, the AI field has been refined and more development has been made in applications such as unsupervised machine learning, natural language processing (NLP), natural language generation (NLG), reinforcement learning, and neural network learning. These tend to emphasise assisting human professionals, so that they can perform at higher levels, accomplishing results that previously were not possible, rather than taking on the work of an entire class of co-workers.

Concerns in this area of job replacement come at a time when researchers are predicting as much as a 98 per cent chance of a global recession in 2023.¹ In this changing environment, the benefits of AI go beyond replacing static, repetitive jobs – such as assembly line workers – with robotics. An inevitable result is that there will be a cutback in the employee workforce in areas most affected by the recession, while at the same time greatly enhancing opportunities for innovation and out-of-the-box thinking. In researching the COVID-19 related downturn and what can be expected thereafter, the Brookings Institution reported:

As for what all of this means for the future, the potential of an automation surge reinforces the fact that any coming recession won’t only bring an end to the nation’s plentiful supply of jobs. Any downturn is likely to bring a new bout of structural change in the labour market and its demand for skills. If it extends for a while, the downturn could induce firms in food service, retail, and administrative work to restructure their operations toward greater use of technology and higher-skilled workers. For America’s beleaguered lower-skill workers, these changes will complicate the return to normalcy.⁶

AI developers are working feverishly to meet anticipated needs to stabilise or reduce costs for business and industry. Artificial intelligence has advanced in ways that were unanticipated by most observers in the twentieth century. As reported in the US Department of Labor Monthly Labor Review for July 2022:

New computing capacities – in areas such as image recognition, robotic manipulation, text processing, natural-language processing, and pattern recognition, and, more generally, the ability to learn and improve rapidly in relatively autonomous ways – represent a break from the hand-coded, rules-based programs of the past. In this view, newer robots and AI represent a clear departure from previous waves of computing, one that accelerates the pace of technological change and job displacement.²

I must admit that I had anticipated that the rapid growth of sophisticated analytical and predictive AI capabilities would be the primary workforce impact in the near term. Driven by the massive data collection of the twenty-first century, we are ripe for harvesting these data in useful ways to drive higher profits and greater efficiencies in operations and logistics. However, most surprising to me is the burgeoning growth of the creative applications of AI in the areas of original copywriting, graphics, and art. These creative, rather than purely analytical, applications seemed further down the road for AI, yet they already have a place in business and industry that is growing stronger every day.

Among the technologies that have grown are those driven by Generative Pre-trained Transformer 3 (GPT-3). Developed by Open AI, that rather ambiguous title covers a range of functional tools that, among other talents, conduct research gathering information from across the internet without supervision and “write” amazingly cogent copy that can be used in news releases, product promotions, operation manuals, and even newspaper and periodical articles. In writing an article¹⁰ on the topic of AI in higher education, I sampled a number of platforms using GPT-3. After describing my goals for the article and filling in a very brief outline, I submitted it to https://copy.ai. A 10-paragraph, nearly 700-word article appeared in just a few seconds. I saved a few sentences of the text written by Copy.ai as an example of the writing:

The world we live in is already being reshaped by artificial intelligence. The technology is changing the way we work, learn and interact with each other, but it’s also perpetuating inequality. As institutions of higher education that are tasked with preparing students for the future of work, colleges and universities have a unique opportunity to help shape this future to be more equitable.

I had not mentioned in my brief submission to the Copy.ai program anything about inequality or the responsibility of educators to ensure that learners are prepared for the less obvious AI ramifications for equity within the workplace. AI picked up on that omission in a second, literally. The remainder of the 10-paragraph essay was equally lucid and cogent.¹⁰ News publications, including Bloomberg and Associated Press among many others, regularly use AI to produce sports and news reports.⁸ Almira Osmanovic Thunström, writing in the Scientific American, reported that she has submitted a GPT-3-written academic article to peer-review.¹¹ Clearly, momentum in original AI research and writing of articles is building, and AI’s role in original writing is already proven.

Specific to business uses, natural language processing (NLP) is a function of AI that enables it to create copy. It powers a host of applications. There are now dozens of AI copywriting tools that promise to do everything from optimising your content for SEO to writing creative product descriptions, to creating video scripts and social media posts.⁵

In business applications, given a modest set of instructions by humans, the algorithm goes about gathering results from selected databases or the internet at large and, within seconds, it assembles multiple sample products that can be refined and disseminated. This ability to “create” truly original products, from text to multimedia to art, is at the forefront of AI today. No longer about robots replacing assembly-line workers, the emerging frontier is much more about creativity and visioning.

Perhaps most striking of the emerging, commonly available, AI capabilities is seen in DALL-E 2, developed by Open AI (which also developed GPT-3). Using brief text descriptions, the algorithm instantly creates stunning original, creative artwork.⁷ As a society, we have yet to fully catch up with the current state of AI. The New York Times reports on some of the challenges presented by the advent of AI in competitions, such as the case in which an AI-created artwork was submitted in a competition. It took first place in an art competition in Colorado; not all of the others in the competition were happy about competing with AI. ⁹

While the images created by AI are stunning, perhaps even more useful day to day is the advent of AI unique coding applications from simple descriptions. Text-to-code generation is now offered by a host of vendors featuring coding on demand by describing the outcome of what you require to be coded. No knowledge of the computer language or programming approach is required; one simply describes what you want the new app to do, and AI writes the code.⁴ The iterative quality of AI-coding AI applications that can in turn code even more unique AI apps is a bit mind-boggling to me.

Where do these innovations lead? Certainly, today they are providing dramatic new ways to cut costs, shorten timelines, and enhance productivity while creating new product lines. Over time, as the algorithms improve and expand, we will see whole new approaches to the creative research and development areas of the economy. The creation of original text and graphics has already impacted marketing, media, and related fields. The coding assistance and creation models will enable us to bridge the disconnect we often find between the creative vision and the coding reality. The outcome, too often, does not today precisely match the vision. The result of this direct connection between person and machine may be a much tighter relationship between the creative visioning and the production side of operations.

With open models leading the way, we are likely to see these new technologies adopted and utilised even under the threatening recession. They offer a bright light illuminating an unprecedented future surge of innovation and responsive production.

This article was originally published on 31 January 2023. 

About the Author

Ray Schroeder is Senior Fellow at the University Professional and Continuing Education Association (UPCEA) and Professor Emeritus at the University of Illinois Springfield (UIS). A frequent speaker and author, Ray writes the biweekly “Online: Trending Now” column in Inside Higher Education and distributes multiple daily curated reading lists for UPCEA.

References

1. Egan, M. (28 September 2022). “There’s a 98% chance of a global recession, research firm warns”. CNN. https://www.cnn.com/2022/09/28/economy/recession-global-economy/index.html
2. 2Handel, M.J. (13 July 2022). “Growth trends for selected occupations considered at risk from automation”. Bls.gov. https://www.bls.gov/opub/mlr/2022/article/growth-trends-for-selected-occupations-considered-at-risk-from-automation.htm
3. Holzer, H.J. (9 March 2022). “Understanding the impact of automation on workers, jobs, and wages”. Brookings. Retrieved 30 October 2022, from https://www.brookings.edu/blog/up-front/2022/01/19/understanding-the-impact-of-automation-on-workers-jobs-and-wages/
4. Janakiram, M.S.V. (14 March 2022). “5 AI tools that can generate code to help programmers”. Forbes. https://www.forbes.com/sites/janakirammsv/2022/03/14/5-ai-tools-that-can-generate-code-to-help-programmers/?sh=24d0449f5ee0
5. Liza. (26 February 2022). “5 best AI copywriting tools in 2022 [out of the 11 we tested]”. P2P Marketing. https://peertopeermarketing.co/ai-copywriting-tools/
6. Muro, M., Maxim, R., & Whiton, J. (24 March 2020). “The robots are ready as the COVID-19 recession spreads”. Brookings. https://www.brookings.edu/blog/the-avenue/2020/03/24/the-robots-are-ready-as-the-covid-19-recession-spreads/
7. OpenAI. (14 April 2022). Dall·e 2. OpenAI. https://openai.com/dall-e-2/
8. Peiser, J. (5 February 2019). “The rise of the robot reporter”. New York Times. https://www.nytimes.com/2019/02/05/business/media/artificial-intelligence-journalism-robots.html
9. Roose, K. (2 September 2022). “An A.I.-generated picture won an art prize. Artists aren’t happy”. New York Times. https://www.nytimes.com/2022/09/02/technology/ai-artificial-intelligence-artists.html
10. Schroeder, R. (24 August 2022). “Higher ed, meet GPT-3: We will never be the same!”. Insidehighered.com. Retrieved 3 November 2022, from https://www.insidehighered.com/digital-learning/blogs/online-trending-now/higher-ed-meet-gpt-3-we-will-never-be-same
11. Thunström, A.O. (30 June 2022). “We asked GPT-3 to write an academic paper about itself—then we tried to get it published”. Scientific American. https://www.scientificamerican.com/article/we-asked-gpt-3-to-write-an-academic-paper-about-itself-mdash-then-we-tried-to-get-it-published/