The 2023 AI Index Report was released in early April by the Stanford Institute for Human-Centered Artificial Intelligence, it’s sixth annual analysis of the impact, progress, and trends of AI. Led by an interdisciplinary group of experts from across academia and industry, the report aims to be the world’s most authoritative source for data and insights about AI.
Last year’s report concluded that 2021 was the year that AI went from an emerging technology to a mature technology, no longer just a subject of speculative lab research but something with real-world economic impact, both positive and negative.
“AI has moved into its era of deployment; throughout 2022 and the beginning of 2023, new large-scale AI models have been released every month,” wrote the report’s co-directors Jack Clark and Ray Perrault about the 2023 report. “Policymakers are talking about AI more than ever before. Industry leaders that have integrated AI into their businesses are seeing tangible cost and revenue benefits. The number of AI publications and collaborations continues to increase. And the public is forming sharper opinions about AI and which elements they like or dislike.”
At over 380 pages, the highly comprehensive report starts out by summarizing its ten top takeaways:
● Industry leads the development of state-of-the-art AI models. “Until 2014, most significant machine learning models were released by academia. Since then, industry has taken over. In 2022, there were 32 significant industry-produced machine learning models compared to just three produced by academia.”
● Performance on traditional AI benchmarks is saturating. Over the past decade, benchmarks have helped track AI’s progress in key application areas including language,vision, and speech. Major AI technical benchmarks are now approaching 80%-90% accuracy, reaching a point where it’s hard to do much better. The median improvement since benchmarks were launched is 42.4%, but the median improvement within the last year is only 4%.
● AI is both harming and potentially helping the environment. Training large AI models can have serious environmental implications, but recent research suggests that AI systems can be used to optimize the energy consumption of large data centers.
● AI models are being used to aid scientific discoveries. In 2022 AI was applied to the study of hydrogen fusion, the efficiency of matrix manipulation, and the generation of new antibodies. In addition, a team at Nvidia trained an AI system to help design smaller, faster, and more efficient chips than traditional EDA tools.
● Incidents and controversies concerning the misuse of AI are rapidly rising. The number of ethical misuse incidents of AI has increased 26 times since 2012, evidence of both greater use of AI technologies and awareness of misuse possibilities. While large generative models are very capable, they also come with ethical challenges. “Chatbots like ChatGPT can be tricked into serving nefarious aims.”
● The demand for AI-related professional skills is increasing across virtually every American industrial sector. “The number of AI- related job postings has increased on average from 1.7% in 2021 to 1.9% in 2022.”
● For the first time in the last decade, year-over-year private investment in AI decreased. “In 2022 the amount of private investment in AI was 18 times greater than it was in 2013. However, “AI private investment was $91.9 billion in 2022, which represented a 26.7% decrease since 2021.” The total number of AI funding events decreased with the exception of the largest ones at over $1 billion.
● While the number of companies adopting AI has plateaued, the companies that have adopted AI continue to pull ahead. “The proportion of companies adopting AI in 2022 has more than doubled since 2017, though it has plateaued in recent years.”
● Policymaker interest in AI is on the rise. “The number of bills containing artificial intelligence that were passed into law grew from just 1 in 2016 to 37 in 2022. An analysis of the parliamentary records on AI in 81 countries likewise shows that mentions of AI in global legislative proceedings have increased nearly 6.5 times since 2016.”
● Chinese citizens are among those who feel the most positively about AI products and services. Americans ... not so much. “78% of Chinese respondents (the highest proportion of surveyed countries) agreed with the statement that products and services using AI have more benefits than drawbacks,” but only 35% of sampled Americans (among the lowest of surveyed countries) agreed with the statement.
Let me briefly discuss a few of the report’s key findings.
Large language models (LLMs) are getting bigger and more expensive. Given that state-of-the-art LLMs require very large amounts of data, compute power, and financial resources, it’s not surprising that industry has been racing way ahead of academia and non profits. For example, GPT-2, — generally considered the first LLM, — was released by OpenAI in 2019 with 1.5 billion parameters in its neural network. GPT-3, initially released in 2020, has 175 billion parameters. GPT-4 was released in March of 2023, and while OpenAI did not specify the model's dimensions, some estimates put its size at 100 trillion parameters.
At a recent MIT conference, OpenAI’s CEO Sam Altman said that the age of giant AI models is over, adding that “the research strategy that birthed ChatGPT is played out and future strides in artificial intelligence will require new ideas.” Further progress in AI will not come from the development of ever bigger AI models, he said, — “we’ll make them better in other ways.”
While leading edge firms pull ahead, the majority are still in the early stages of deployment. “The heightening integration of AI and the economy comes with both excitement and concern,” notes the report while raising a number of important questions. “Will AI increase productivity or be a dud? Will it boost wages or lead to the widespread replacement of workers? To what degree are businesses embracing new AI technologies and willing to hire AI-skilled workers? How has investment in AI changed over time, and what particular industries, regions, and fields of AI have attracted the greatest amount of investor interest?”
Leading edge firms are placing AI at the center of their business strategies. But a number of recent surveys show that the majority of enterprises are still in the early stages of AI deployment and risk being left further behind. For example, a McKinsey survey on the state of AI in 2022 found that while adoption has more than doubled since 2017, the proportion of organizations using AI has leveled off at 50% to 60% in recent years. Organizations that lead in embracing AI report meaningful revenue and productivity increases and continue to pull ahead of competitors.
Similarly, a 2022 Deloitte survey of 2,800 executives from advanced economies found that 28% of respondents were deploying AI at scale and achieving high outcomes, but 46% were still in the early stages of deployment with no significant outcomes. And, in another recent survey of over 1,600 C-suite executives of the world’s largest companies, Accenture found that only 12% had the strategic and operational AI capabilities needed to achieve superior growth, while the majority of firms, 63%, were still at the experimenting stage and had only average AI capabilities.
Once again, the United States leads in total AI private investment. “In 2022, the $47.4 billion invested in the United States was roughly 3.5 times the amount invested in the next highest country, China ($13.4 billion), and 11 times the amount invested in the United Kingdom ($4.4 billion).” In addition, the US led in total number of newly fund AI companies at 542, followed by China (160), and the UK (99).
The industry sector that attracted the most private investment in 2022 was medical and healthcare ($6.1 billion); followed by data management, processing, and cloud ($5.9 billion); fintech ($5.5 billion); cybersecurity and data protection ($5.4 billion); and retail ($4.2 billion). In addition, the most widely deployed AI capabilities in businesses include “robotic process automation (39%), computer vision (34%), NL text understanding (33%), and virtual agents (33%).
The demand for AI professional skills continues to increase across virtually every industry sector. The sectors with the most AI job postings in 2022 were information technologies (5.3%); professional, scientific, and technical services (4.1%); finance and insurance (3.3%); and manufacturing (3.3%).
The increasing demand for such skills is driving more AI specialization. “The proportion of new computer science PhD graduates from U.S. universities who specialized in AI jumped to 19.1% in 2021, from 14.9% in 2020 and 10.2% in 2010.”
Not surprisingly, new AI PhDs are increasingly heading to industry. In 2011, 40.9% of new AI PhDs took jobs in industry, roughly the same proportion, 41.6%, that took jobs in academia. But in 2021, 65.4% of new AI PhDs went to industry, more than double the 28.2% who took jobs in academia.
“AI will continue to improve and, as such, become a greater part of all our lives,” wrote AI Index co-directors Jack Clark and Ray Perrault in conclusion. “Given the increased presence of this technology and its potential for massive disruption, we should all begin thinking more critically about how exactly we want AI to be developed and deployed. We should also ask questions about who is deploying it — as our analysis shows, AI is increasingly defined by the actions of a small set of private sector actors, rather than a broader range of societal actors.”
Comments