AI’s Impact on the Economy and Employment
My friend Peter Christy is doing research on how AI is impacting / will impact employment and wants people’s feedback (and if you have some thoughts, please respond here). This reminded me that I had written on this very topic in a section that was to be in the chapter on AI in my book Containing Big Tech. In the end, I cut this section out of the chapter on AI (the chapter was getting crazy long), with some of the themes from that section moving into the chapter on Competition and Antitrust. But I figured I would revive this section from the cutting room floor and publish it in my blog because it is relevant to Peter’s query (note that I also used parts of that section in this editorial here). I think my main feedback to Peter’s query is what I wrote as a comment on LinkedIn:
“I think the big concern with AI is that unlike past issues with automation and loss of jobs/displacements, in this iteration, the technology (AI) is actually pretty consolidated/concentrated with a handful of Big Tech vendors (with prior tech revolutions, we did not have 3-4 companies doing all the weaving machines or operating all the railways on a worldwide basis), so you also need to factor in the impact of the large tech monopolies that emerged at the end of the current 3rd industrial revolution leveraging their large collection of data and computing horsepower to take over and dominate the 4th industrial revolution. The point is it is not just about the impact of this new tech on the economy/employment, as it was in the past waves of automation, but it is also the impact of new tech (AI in this case) AND tech monopolies/concentration of power on the economy/employment.”
So here is the section on this topic that was cut from my book that touches on Peter’s research.
<beginning of section>
AI and the Economy
Much of the discussion in this chapter has been around how AI systems can introduce bias that can lead to discrimination and the proposed legislation in Europe and the US that can mitigate this. But before I close out this chapter, I want to step back even further and talk about the potential impact of AI on the economy and market competition.
As discussed previously, AI has many benefits, including increasing labor productivity by automating routine tasks to free workers to focus on higher-level tasks. Thus, AI has the promise to let us do more with fewer resources, which in theory would enlarge the economy. But, as noted by Economist Joseph E. Stiglitz, what we have seen is that over the last four decades is that almost all of the “gains of increased productivity have gone to the top, while the middle has stagnated and many at the bottom have seen both incomes and opportunity wane.”[1]
It is unclear what impact AI will have on this decades-long stagnation of wages. However, it has been estimated that 50 to 70% of this languishing over the last four decades can be mainly attributed to the loss of routine jobs to automation.[2] The rise of AI significantly furthers this automation trend, given AI’s ability to handle even more complex tasks, including those that would typically be handled by more skilled workers that were historically not impacted in prior decades’ moves to automate. For example, McKinsey estimated in 2017 that 50 percent of current work activities are automatable, and 6 out of 10 current occupations have more than 30 percent of their work activities that could be automated. Furthermore, McKinsey estimated that worldwide, between 400 to 800 million individuals could be displaced by automation and would need to find new jobs by 2030.[3]
That being said, some economists are projecting a net increase in jobs with the rise of AI, but there will be significant worker displacement. For example, the World Economic Forum estimated in 2020 that by 2025 over 85 million jobs would be displaced by AI, but 97 million new jobs would be created by AI across 26 countries, meaning there will be a net increase of 12 million jobs.[4] That is the potential upside. But the downside is that this difficult displacement process could include lower wages for even more people, further accelerating inequality. And given the societal upheaval that we have seen over the last few decades due to rising inequality and globalization that in turn has fueled authoritarianism and polarization, the growth of AI and its ability to accelerate the displacement of jobs may further test our democracy.[5] Governments, schools, and employers will need to acknowledge this issue and step up to help workers ease through this transition.
Specific to market competition, Big Tech firms are certainly taking AI very seriously — Google CEO Sundar Pichai declared that AI will “have a more profound impact on humanity than fire, electricity, and the internet.”[6] Big Tech firms, including Amazon, Alphabet/Google, Meta/Facebook, and Netflix, are now responsible for $2 of every $3 spent on AI. They are also among the largest recipients of US patents for AI. And from 2013-2018, approximately 90% of successful AI startups were purchased by large technology firms.[7] In addition, the median number of AI employees for Amazon, Alphabet, Microsoft, Meta, and Apple is 18,000 compared to 2,500 for the next 20 tech players.[8]
As I will discuss in the chapter on Antitrust and competition, many consider Amazon, Alphabet, Meta, and Apple monopolies in their respective market segments. In light of Salesforce CEO Marc Benioff’s comments that AI is fueled by data and that the more data you train the AI system, the better its capabilities, the fact that Big Tech companies collect massive amounts of data and also have the world’s largest computing infrastructures further gives them a significant competitive advantage when it comes to AI. Therefore, this massive investment in AI by Big Tech, combined with their “hoard” of data and their elastic cloud-based computing environments, lends itself to these same Big Tech firms cornering the market for AI and automation.[9]
This consolidation may not favor innovation as we embark on the “Fourth Industrial Revolution” — the fusion of advances in AI, robotics, IoT, etc., that blur the lines between the physical, digital, and biological spheres.[10] This consolidation and lack of innovation may further yield greater societal inequality.
<end of section>
Possible Mitigation Solutions
In the aforementioned editorial, I did throw out some misc. ideas and suggestions, so I thought I would also include those here:
“Governments, schools, and employers must acknowledge this issue and step up to help workers ease through this transition. We need to be smart about this, as we can’t say let’s retrain everyone to be software developers, as AI is increasingly being built to automate the development of software programs.
…
As regulators and lawmakers look at antitrust issues concerning Big Tech firms, AI should not be overlooked. To paraphrase Wayne Gretzky, regulators need to skate where the puck is going, not where it has been. AI is where the puck is going in technology. Therefore, acquisitions of AI companies by Big Tech companies should be more closely scrutinized. In addition, the government should consider mandating open intellectual property for AI. For example, this could be modeled on the 1956 federal consent decree with Bell that required Bell to license all its patents royalty-free to other businesses. This led to incredible innovations such as the transistor, the solar cell, and the laser. Thus, when policymakers look at AI, it is equally critical to address whether or not it is healthy for our economy to have the equivalent of electricity concentrated in a few firms’ hands.”
I did not further explore this area and/or research possible solutions to address displacement (my chapter on AI covered many other AI-related issues with corresponding recommendations, so my plate was full, so another reason this got cut). Hopefully, Peter’s eventual analysis can provide some recommendations and guidance.
FOOTNOTES
[1] Julia Angwin, “How AI Could Undermine an Efficient Market Economy,” The Markup, June 25, 2022, https://themarkup.org/newsletter/hello-world/how-ai-could-undermine-an-efficient-market-economy.
[2] Daron Acemoglu and Pascual Restrepo, "Tasks, Automation, and the Rise in US Wage Inequality," National Bureau of Economic Research, June 2021, https://www.nber.org/papers/w28920.
[3] McKinsey, Jobs lost, jobs gained: What the future of work will mean for jobs, skills, and wages," November 28, 2017, https://www.mckinsey.com/featured-insights/future-of-work/jobs-lost-jobs-gained-what-the-future-of-work-will-mean-for-jobs-skills-and-wages.
[4] Mohamed Kand and Murat Sonmez, "Don't fear AI. It will lead to long-term job growth," World Economic Forum, October 26, 2020, https://www.weforum.org/agenda/2020/10/dont-fear-ai-it-will-lead-to-long-term-job-growth/.
[5] Julia Angwin, “How AI Could Undermine an Efficient Market Economy,” The Markup, June 25, 2022.
[6] Tom Knowles, “AI will have a bigger impact than fire, says Google boss Sundar Pichai,” The Times, July 13, 2021, https://www.thetimes.co.uk/article/ai-will-have-a-bigger-impact-than-fire-says-google-boss-sundar-pichai-rk8bdst7r.
[7] Dakota Foster, “Antitrust investigations have deep implications for AI and national security,” The Brookings Institution, June 2, 2020, https://www.brookings.edu/techstream/antitrust-investigations-have-deep-implications-for-ai-and-national-security/.
[8] Bhaskar Chakravorti, "Biden’s ‘Antitrust Revolution’ Overlooks AI—at Americans’ Peril," Wired, July 27, 2021, https://www.wired.com/story/opinion-bidens-antitrust-revolution-overlooks-ai-at-americans-peril/.
[9] Bhaskar Chakravorti, "Biden’s ‘Antitrust Revolution’ Overlooks AI—at Americans’ Peril," Wired, July 27, 2021.
[10] Klaus Schwab, “The Fourth Industrial Revolution: what it means, how to respond,” World Economic Forum, January 14, 2016, https://www.weforum.org/agenda/2016/01/the-fourth-industrial-revolution-what-it-means-and-how-to-respond/.