🧠 GenAI Tradeoffs: Efficiency vs. Efficacy

Subscribe | AI Toolkit | Meet The Team

Welcome back AI prodigies!

In today’s Sunday Special:

• 🤖Technological Progress and Automation

• 👟Speed Matters

• 💥Essential Factors of Disruptive Innovation

• 🔑Key Takeaway

Read Time: 7 minutes

🎓Key Terms

• Spinning Jenny: a hand-powered machine capable of spinning multiple threads simultaneously, revolutionizing textile production.

• Generative AI (GenAI): a technique that learns patterns from existing datasets to generate new content, such as text, images, audio, video, and code.

🩺 PULSE CHECK

🤖TECHNOLOGICAL PROGRESS AND AUTOMATION

Earlier this year, The Daily Show host Jon Stewart addressed the Generative AI (GenAI) revolution. In his playful style, he criticized Big Tech executives for downplaying the potential job loss it could create. All technological shifts—agricultural, industrial, and computer—disrupted labor markets. However, no one would argue that they weren’t worthwhile.

For most of human history, roughly 90% of people were farmers. As recently as 1840, 75% of American workers farmed. The few who didn’t were craftspeople like blacksmiths, woodworkers, or tailors (i.e., aside from women who were confined to household labor). As the first Industrial Revolution slowed down by 1850, the Spinning Jenny had already revolutionized the textile industry by enabling the mass production of yarn at a lower cost. As a result, it replaced textile producers. Steam-powered trains replaced horse-and-buggy options (i.e., carriages) because of speed, efficiency, and capacity. Mechanized iron production with a blast furnace, rolling mill, and forging press substituted most blacksmiths because of speed, efficiency, and consistency. By 1900, just 35% of Americans worked in agriculture.

Scientific breakthroughs and technological innovation enable us to perform high-order work; it always has. Like every revolution, future jobs are less evident in the early innings of the cycle. But the odds are, uniquely human skills like empathy, ethical decision-making, creativity, and managing people will become far more critical as GenAI works its way through the economy. Like other technological revolutions, AI adoption will likely produce other consequences beyond job automation that are difficult to predict and nearly impossible to prevent.

👟SPEED MATTERS

The speed of adoption will determine the extent of disruption. Consider the automobile; German inventor Carl Benz patented it in 1886. Roughly 15 years later, there were only 8,000 automobiles in America. It quickly grew to 500,000 automobiles by 1910. Still, that’s 25 years, and even then, only about 1 in 200 Americans had an automobile. The first stop sign wasn’t employed until 1915 (i.e., although traffic lights at major intersections came slightly earlier). And car ownership didn’t reach 50% until 1950. This pace gave us time to sort out formal regulations and societal norms. Of course, software usage can occur exponentially.

Social media had negligible usage until 2008, when Facebook went from a few million users to a billion accounts in just four years. Social media has subsequently been shown to cause cyberbullying, self-esteem issues, body image issues, depression, and a host of other mental health problems, not to mention the vast scale spread of misinformation. We were well past a billion active users before there was any data on social media’s risks and negative impact. In contrast, the risks were less subtle with automobiles, allowing the public to take preventative measures, like putting in stop signs or requiring driver’s licenses, before it scaled up and made the problem more unmanageable.

Unlike social networks, which took years to achieve widespread usage, AI adoption is happening much faster. OpenAI’s ChatGPT was the fastest-growing digital product of all time, reaching 100 million users in just two months. No one reads product safety warnings; we don’t even have safety warnings for AI. Aside from accuracy disclaimers, we don’t fully understand it yet.

💥ESSENTIAL FACTORS OF DISRUPTIVE INNOVATION

In the fictional world of the science fiction (i.e., “sci-fi”) franchise Star Trek, a space force called Starfleet had a Prime Directive. This regulation was implemented to protect less mature species from harming themselves with technology that is more advanced than it was. It outlined three essential factors behind a technology’s disruptive capacity:

1. Adoption Rate: the rate at which society adopts a new technology.

2. Impact Radius: a conceptual framework for a technology’s total impact on society. It’s calculated as (Size of Impact) x (Number of People Impacted) x (Duration of Impact).

3. Learning Curve: the rate at which society learns to use technology to achieve its goals and subsequently understand its impact.

Note that the impact is real regardless of where we’re at on the learning curve. People were suffering body image problems from social media as it became widespread, even if we weren’t yet aware of it.

If the impact radius is small, we can pilot a technology and limit the risks. The space shuttle was a great technology, but it came with risks. We lost some astronauts and money in NASA’s Challenger STS-51L Accident and Columbia STS-107 Mission disaster. While any loss of life is tragic, it was undertaken by a limited number of people who understood those risks.

When the adoption rate exceeds the impact radius (e.g., social media), especially when it’s steeper than our learning curve, we have excessive risk; we create effects faster than we can understand them across vast populations. That’s a recipe for unintended consequences and unpreventable cybersecurity threats to critical infrastructure like energy or telecommunications.

It’s OK to move fast and break things when it’s in your own home. It’s harder to justify that philosophy when you’re barreling down the street in a 1,200-pound Ford Model T car in 1930 at 40 Miles per Hour (MPH). It’s even harder when half the neighborhood is doing it. Unfortunately, producing and publishing content or generating code with AI isn’t confined to your neighborhood.

Looking back at the Industrial Revolution, we moved fast and broke things in the name of innovation. We strip-mined land, deforested mountains, and polluted the water. Famously, Ohio’s Cuyahoga River caught fire as recently as 1969; that’s right, the concentration of chemical pollutants was so high that river water became flammable. In fact, the river caught on fire multiple times, but no one cared. In his book Chemistry of the Environment, environmental chemist David Newton wrote, “Fundamentally, this level of environmental degradation was accepted as a sign of success.” But how much social degradation is acceptable, or even desirable, for progress?

🔑KEY TAKEAWAY

AI automation concerns are legitimate, but the risk of harm itself isn’t a reason not to use it. We know cars kill many people each year, but we feel the benefits outweigh the risks because we can measure the benefits and risks. However, given the opaque nature of most AI models, implementing GenAI into knowledge worker tasks (e.g., a lawyer’s document draft or an accountant’s financial report) is less understood, especially by knowledge workers themselves. Navigating the tradeoff between accuracy and efficiency is critical, especially when the impact radius of implementation touches every industry, company, and product.

📒FINAL NOTE

FEEDBACK

❤️TAIP Review of The Week

REFER & EARN

🎉Your Friends Learn, You Earn!

You currently have 0 referrals, only 1 away from receiving ⚙️Ultimate Prompt Engineering Guide.

Refer 3 friends to learn how to 👷‍♀️Build Custom Versions of OpenAI’s ChatGPT.

Copy and paste this link to friends: https://theaipulse.beehiiv.com/subscribe?ref=PLACEHOLDER