Dear CIO,

In February 1928, the New York Times published an article that could have been written today - just swap "machines" with "artificial intelligence." The piece "March of the Machine Makes Idle Hands" captured a moment of intense technological anxiety as automation transformed American industry. Nearly a century later, we find ourselves in a strikingly similar conversation about AI's impact on the workforce, particularly regarding generative AI and coding tools. In this edition, we are going to look at the parallels between 1928 and today, and how history can help us better navigate through the AI revolution.

Best Regards,
John, Your Enterprise AI Advisor

Everything Old is New Again

Echoes of 1928 in Today's AI Anxiety

The Parallel Fears

In 1928, factory owners were installing machines that could do the work of multiple people. Today, companies are implementing AI systems to generate code, write content, and create designs in minutes. The core fear remains unchanged: will technology make human workers obsolete?

The 1928 article noted that factories were "producing 42 percent more than they did eight years ago with 14 percent fewer workers." Today's statistics about AI productivity gains sound remarkably similar. Studies suggest that developers using GitHub Copilot complete tasks significantly faster, while GPT-4 can match or exceed human performance on various professional tasks.

The Human Element: Then and Now

What's particularly striking is how fundamental concerns about human value persist in an automated world. The 1928 article quotes labor statistics showing increasing output alongside decreasing employment—a pattern that modern workers fear could repeat with AI. Just as factory workers worry about machines replacing their manual labor, today's knowledge workers worry about AI replacing their cognitive labor.

Different Technologies, Same Narrative

The Industrial Revolution primarily automated physical tasks - the mechanical reproduction of human labor. Today's AI revolution targets intellectual tasks - the computational reproduction of human thought. Yet the narrative structure of public concern remains remarkably consistent:

1. Initial excitement about increased productivity

2. Growing concern about job displacement

3. Debates about the role of human workers

4. Questions about the economic distribution of benefits

5. Calls for policy intervention and worker protection

Learning from History

The 1928 article's concerns about widespread unemployment didn't fully materialize as predicted. While individual industries saw significant disruption, new jobs and industries emerged. This historical perspective offers both caution and comfort regarding today's AI fears:

• Caution: The transition period was difficult for many workers, requiring significant adaptation and retraining

• Comfort: Human work didn't disappear; it evolved

Key Differences

Despite the similarities, there are crucial differences between 1928's automation and today's AI revolution:

1. Speed of Change: AI technology is evolving much faster than industrial automation did

2. Scope of Impact: AI potentially affects a broader range of jobs, including high-skill knowledge work

3. Adaptation Potential: Modern workers generally have more access to education and retraining opportunities

4. Economic Context: Today's global digital economy offers more flexibility in creating new types of work

Jevons Paradox: Efficiency Breeds More Demand

While concerns about AI-driven job losses mirror past fears of mechanization, history suggests that increased efficiency often fuels greater demand rather than rendering human labor obsolete. This dynamic, known as Jevons Paradox, was first observed in the 19th century when economist William Stanley Jevons noted that more efficient steam engines led to greater coal consumption, not less. As industries found new ways to leverage cheaper energy, overall coal demand surged.

A similar paradox is emerging with AI. Generative AI tools like GitHub Copilot, ChatGPT, and Midjourney may reduce the effort required for individual tasks, but rather than eliminating work, they often expand the scope of what’s possible. Companies can now produce more software, more content, and more designs at a lower cost, potentially driving an increase in total work output. Instead of replacing programmers, AI-assisted coding might lead to more software development than ever before, as barriers to entry drop and new applications emerge.

The automation wave of the 1920s increased industrial efficiency, but it didn’t halt economic growth or human employment—it reshaped them. Likewise, AI’s greatest impact may not be in making certain jobs redundant but in amplifying the scale and ambition of human creativity and productivity. Understanding this paradox challenges the assumption that AI’s labor-saving potential automatically translates into fewer jobs; instead, history suggests that new efficiencies often generate more opportunities than they eliminate.

The Path Forward

Just as the industrial revolution ultimately led to new forms of work and economic growth, the AI revolution will likely create opportunities we can't yet imagine. However, the 1928 article reminds us that managing technological transitions requires careful attention to their human impact.

Some key lessons for today:

1. Proactive Adaptation: Rather than resisting technological change, focus on preparing for it

2. Skills Evolution: Identify and develop skills that complement rather than compete with AI

3. Policy Consideration: Consider how to ensure the benefits of AI advancement are broadly shared

4. Historical Perspective: Remember that technological disruption, while challenging, often leads to new opportunities

Conclusion

The 1928 New York Times article is a fascinating mirror of our current moment. While the specific technology has changed, the human experience of facing transformative technological change remains remarkably consistent. As we navigate the AI revolution, this historical parallel offers a valuable perspective: technological change is not new, and humanity has adapted before.

The most important lesson is that while technology changes the nature of work, it doesn't eliminate the need for human contribution. As it was in 1928, the challenge is managing the transition in a way that preserves human dignity and economic opportunity while embracing technological progress.

The machines of 1928 didn't end human work; they transformed it. AI will likely do the same. The question isn't whether humans will continue to work but how that work will evolve and how we can ensure that evolution benefits society as a whole.

• George Ftizmaurice highlights concerns that junior software developers are becoming overly reliant on AI coding tools.

• Shaun Nichols reports on MITRE researchers having proposed OCCULT, a new framework for assessing AI-driven cybersecurity threats.

• Florian Maier covers a study that simulated an AI-driven disinformation campaign, revealing that generative AI could effectively manipulate public perception and trigger financial instability.

• Reuven Cohen is doing incredible research on AI and reminds us that less is more.

• The Artificially Intelligent Enterprise looks at how AI-powered models are automating workflows.

• AI Tangle covers Anthropic’s Release of a new Code Maestro Sonnet Model & Dev Tool.

Your AI, Your Rules—Deploy Open-Source LLMs at Scale

Open-source LLMs like IBM’s Granite offer the flexibility and transparency businesses need for real AI innovation. I’m co-hosting this talk to show how enterprises can build AI solutions without giving up control.