How Has COVID Affected the AI Economy?

COVID-19 put incredible strain on healthcare systems worldwide. With so much news revolving around healthcare and the impact AI had on handling the pandemic, not much has been reported about what sort of effects COVID-19 has had on the AI economy. As companies around the world manage ongoing effects of the pandemic and start to envision life beyond it, people are wondering what was the impact of COVID-19 on the AI industry?

Progression of AI in a Post-Pandemic World

COVID has changed many things about our world, including the use of advanced analytics and AI. Overall in 2020, AI hiring, investment, and adoption increased across the board, following the long-term trends in the industry that are expected to outlast the pandemic’s effects. The crisis also necessitated solutions in days, requiring new, rapid approaches to AI models and technologies. AI also needed more agile methods and resilient models that will ensure new solutions to tackle future crises and challenges.

Let’s take a more in-depth look at some of the ways COVID has impacted the AI economy.

Investment in AI

AI-focused private companies are continuing to see a rise in investment, especially in industries linked to COVID. Total worldwide investment in AI grew 40 percent from 2019 to 2020, compared with only about a 12 percent increase the year prior. The biggest increases in AI investment were, predictably, in healthcare and pharmaceuticals. Private companies funneled $13.8 billion into AI drug discovery in 2020, more than any other investment area. But industries ranging from education to retail to automotive also showed increased AI investment, suggesting the change will outlast the pandemic.

Agile Data Science

The spread of the pandemic caught most governments, businesses, and citizens totally off guard. It quickly turned into an economic crisis and continues to be a supply chain crisis. Business leaders needed to act quickly, providing a chance for advanced analytics and AI-based approaches to play a role in decision making. Old machine learning models had a development time of four to eight weeks, at best. But to keep up with things in a COVID economy, minimum viable AI models needed to produce solutions in days, not weeks or months.

Multi-Agent Simulations

The uncertainty produced by the COVID crisis affected everything from healthcare to consumer behavior to the economy as a whole. It also necessitated and expedited adoption of new AI techniques. Because uncertainty produced emotionally driven behavior all over the world, data scientists needed a new framework to evaluate plans and decisions. Scenario analysis became the most prevalent paradigm for monitoring the progression of COVID, economic downturn and recovery, supply chain disruptions, workforce planning, and more. This kind of scenario analysis may have been common in the business world, but using AI simulations to understand all kinds of causal linkages and develop plans of action has been a radical shift in many sectors.

Human behavior modeling will remain an important part of the scenario analysis. This helped model and predict things like how well stay-in-place orders were working. AI techniques were used to perform scenario analyses on daily mobility data, such as how many miles were driven within each zip code in a country, for example. Agent-based simulation and system dynamic modeling are AI techniques that were put to new use in monitoring and predicting COVID-19 disease progression, government interventions, supply disruptions, and more. Both methods have been used in uncertain scenarios to help make strategic and operational management decisions.

AI-Related Hiring

The AI economy also saw a strong uptick in hiring during the pandemic. Across 14 countries analyzed, the AI hiring rate was 2.2 times higher in 2020 than in 2016, on average. Hiring demand going forward will show how prevalent AI is in so many sectors, such as manufacturing, retail, healthcare, finance and other service industries. This means we can expect an increase in demand for AI skills across the board.

Model-Based AI

COVID was such a rare event for so many reasons, one of which is that there was so little historical data about the way a global disease might impact the world. For AI, that meant there was not much information to build model-free approaches to AI, such as deep learning. As a result, model-based AI started to rise again. After some time and the availability of more data, data-rich and model-free approaches could be combined, leading to hybrid solutions. The pandemic certainly served to highlight ways that model-free AI would not be able to function properly.

Countries also need new economic growth models and 5G-powered AI is at the forefront as everyone prepares for what’s next post COVID. The recession is driving the adoption of more business models powered by AI, too. Over 80 percent of the daily moves in the U.S. stock market are now believed to be AI machine-led algorithm trading models.

Human and AI Collaboration

In a new world of work, learning to figure out what machines can do — as well as what they can’t, or shouldn’t, do — will be a critical aspect of adopting more AI technology for workers and CEOs alike. In addition, there will be a focus on human strengths and capabilities, both as they stand on their own and because of the need to integrate them into AI systems.

Conclusion

As companies manage the end of the pandemic and look ahead into a post-pandemic world, data scientists and AI scientists are putting advanced techniques and tools to use to make business decisions. Data science and advanced analytics will start to inform various parts of strategic and operational aspects of business in new ways. AI is slated to have a strong and lasting impact throughout these transitions, and companies that most readily adopt these new technologies will be in the best position to succeed.