7 ways a recession could impact AI and ML in 2023

What would a 2023 recession mean for the artificial intelligence (AI) and machine learning (ML) sector?

Dana Peterson, chief economist at the Conference Board, told CNBC this week that 98% of the CEOs it surveyed are preparing for a recession, up from 95% earlier this year. And Bank of America strategists said last Friday the US could fall into a recession over the next 10 to 12 weeks. 

How would that affect both users of AI and the vendors that supply AI tools and expertise? Here are seven important ways a recession could impact the sector:

1. Well-defined use cases will be key.

A  recession may have a negative impact on the AI workforce in the short term, but there are some AI-driven use cases that will see faster growth and adoption, according to Artem Kroupenev, VP of strategy at machine performance provider Augury. 

“In the industrial AI space, we are seeing a narrower focus on solutions that are showing fast and concrete value within well-defined use cases and that are easy to adapt by non-expert users,” he said. “At the same time, we’re seeing pullback on solutions where the value is not clear, or the use case is yet to be well-defined.” 

Machine health, or predictive maintenance, is a good example of a well-defined industrial use case, he explained: “The combination of AI and IoT drives fast and dramatic operational improvements and does not require any significant change of user behavior in order to adopt it at scale.”

2. Enterprises that already use AI will reap the benefits.

If COVID was any indicator, there will be two scenarios for enterprise use of AI during a recession, according to Wayne Butterfield, partner, ISG Automation, a unit of global technology research and advisory firm ISG. The ‘already haves,’ he explained, will continue to reap the benefits of their previous AI investments, maintaining or expanding the cost savings they already enjoy while using the technology to improve customer and employee experience, gain competitive advantage and grow their top line.

 Those that ‘have not,’ on the other hand, will continue to tread carefully. “In the best of times, it’s a bold move to bet big on an unproven technology,” Butterfield said in an email. “So, unless AI is already delivering results, it will take a brave executive to double down now. Instead, executives will turn to other tried-and-true methods of cost reduction.” 

Of course, the use of AI should not be considered purely a cost-out play. Indeed, automation and AI have never been more important, given the skills and labor shortage many industries are facing. Many organizations will struggle to deliver even on the basics without AI technology augmenting their already-stretched workforces.

3. Adoption of new tech like generative AI will accelerate.

According to Noam Fine, head of AI at telecommunications company Vonage, a global recession is an accelerator for the adoption of new technologies. 

“Enterprises are gaining new trust and excitement about AI, and specifically conversational AI, through new generative AI services built on large data sets,” he said. “This will lead to pilots and willingness to experiment with new solutions as a way of better managing the new economic situation.” The result, he added, is that companies in the space “will find a new open door that wasn’t there before.” 

He pointed out that these solutions will be proven during 2023, getting preliminary traction, but he doesn’t see this as a driver of layoffs in the sector. “I believe we will exit 2023 with new technologies in place, new investment into the workforce to support such technologies, and with a quickly growing AI-based line of businesses for companies in the space,” he said. “As we get closer to 2024 and with further successes in this space, employee resources will be shifted to educate and train employees to handle and manage these new services.”

4. Organizations could be forced to lean on AI more than ever.

Contrary to what one might believe, a 2023 recession could force organizations to lean on technology more than ever, resulting in the AI landscape rapidly expanding, predicts David Raissipour, chief technology and product officer at cybersecurity service company Mimecast. 

“With the likelihood of under-resourced teams that are already fighting for talent – like cybersecurity – businesses will likely look to implement AI solutions to augment business critical operations, including driving cyber protection,” he said. “Bad actors tend to feed on economic uncertainty, with the understanding that there are minimal resources and increased human error during these times.” In addition, tough budgetary environment where organizational leadership must make decisions about the cyber solutions that address the most critical needs, results in an organization that is more vulnerable to cyberattacks.

New AI technology should be expected to be developed in an effort to solve an array of potential cyberattacks, he added, while mitigating the detrimental business risks posed by a recession.  “There’s an opportunity to develop AI technology that can dig deeper – whether that’s a single email or a chain of communications – to understand social graphs and metadata, thus enriching algorithms to better identify risks.” 

5. Laid-off ML talent will trickle into startups.

Moses Guttmann, CEO and co-founder of MLops platform ClearML, says recent layoffs in machine learning are likely the most recent hires as opposed to the more long-term staff that have been working with ML for years. “Since ML and AI have become a more common technology in the last decade, many big tech companies began hiring these types of workers because they could handle the financial cost and keep them away from competitors – not necessarily because they were needed,” he said. From this perspective, it’s not surprising to see so many ML workers being laid off considering the surplus within larger companies.” 

However, as the era of ML talent hoarding ends, it could usher in a new wave of innovation and opportunity for startups, he explained. “With so much talent now looking for work, we will likely see many of these folks trickle out of big tech and into small and medium-sized businesses or startups,” he said. 

However, to fill the void of fewer people on deeply technical teams, companies will have to lean even further into automation to keep productivity up and ensure projects reach completion. “We expect to also see companies that use ML technology put more systems into place to monitor and govern performance and make more data-driven decisions on how to manage ML or data science teams,” he said. 

6. The cost of running AI will become a big focus.

Currently, AI is incredibly expensive and consumes massive amounts of energy. According to Dr. Vishal Sikka, CEO and founder of systems software company Vianai, to prepare for a potential recession organizations need to seek to bring down the cost and improve the acceleration of AI performance by “many orders of magnitude.”  

Only then, he explained, can AI really become effective and enter our everyday lives. “One of the best places for enterprises and vendors to start is to look at the tools for optimizing AI performance,” he said. “Leaders should examine their tech stack and determine which platforms offer the best ROI and how they support employee work with AI’s help.” 

Organizing these tools, he added, will benefit companies making budget cuts. “It reduces overall infrastructure cost in a turbulent economy,” he said. 

7. Companies will invest in AI projects with direct revenue impact.

If a 2023 recession were to occur, in order to maximize their bottom line, enterprises will be more inclined to invest in AI projects that have an immediate and direct impact on revenue generation — rather than taking the risk of investing in long-term fundamental research, said Plamen Minev, technical director, AI and cloud at IT services company Quantum. “This will lead to restructuring and staff reduction, which we are already witnessing,” he said. 

At the same time, businesses will realize the potential of AI and take advantage of its benefits, he added, pointing out that in the last couple of years, AI technology has rapidly evolved and become more accessible and practical. 

By implementing AI into areas such as software design and development, document review management, medical diagnosis and drug discovery, “professional staff can skyrocket their productivity like never before.”