Beyond the Hype: Where the Real Value of AI Lies (And Why You Should Stop Doing Robot Work)
The conversation between Matteo Stroll and the host dives deep into the future of work, the often-misunderstood value of Artificial Intelligence, and the essential steps businesses must take to truly capitalize on this technological age.The Robot Revolution Isn’t Coming for You—It’s Coming for Tedious Tasks
A core theme of this discussion is the necessary shift in perspective regarding automation. Humans are excellent problem solvers, but we are prone to error—miscalculating, mistyping, or making mistakes. As Matteo Stroll notes, “Robots aren’t coming for your jobs. They’re coming for theirs”.
We must stop performing the work of robots. Historical evidence supports this: automation in the auto industry in the 1980s did not destroy humanity; it increased the quality and lifespan of vehicles by scaling human efforts and taking out the errors. The lesson is clear: computers and AI are best viewed as an augmentation and acceleration of human capabilities.The Real Power of AI Is “Unsexy”
While the current media obsession is with generative AI (like creating wallpapers, graphics, or resumes), the speakers argue these user interfaces are “flawed”. The fact that over two-thirds of the world’s adult population is accessing the internet through billions of devices contributes to the “lowest common denominator responses we’re getting out of GPT”.
The true, deep power and value of AI—the kind that will impact society for decades—is the “unsexiest” side, the stuff you won’t see on Bloomberg TV. Fundamentally, for the public and premium users, AI performs three core functions: it organizes, scales, or accelerates.Implementing AI: Define the Problem, Find the Value
For executives seeking real return on investment (ROI), the implementation must be strategic, not ego-driven.
1. Document Your Processes First.
Before attempting to scale or automate, you must document your existing processes. Without this foundation, efforts will yield short, shallow gains with little true value, even if investors want to hear about AI adoption.
2. Hand Over the Tasks.
The key is simple: identify the tasks humans are doing, determine opportunities to automate, and then “get out of the way” to allow the AI to perform. The value metric is the time saved by automating the process.
3. Success Case: The Airline Automation.
Matteo shared a compelling example of an airline with systems administrators spending excessive time on patching 56,000 server devices.
-
A digital solution was created in just six weeks.
-
It automated 98.5% of the patching work without failure.
-
The automation reduced self-inflicted security tickets by 3,000 to 5,000, allowing the security team to focus on actual threats and reducing necessary conversations with federal regulators.
-
Crucially, it improved the quality of life for the systems administrators, who could now focus on their real jobs: managing exceptions.
4. Change Management and Mindset.
Despite the benefits, the systems manager and IT director initially “fought” the change because they feared being replaced. Leaders must remember that business resiliency and competitive nimleness are far more valuable for a public company than focusing solely on headcount reduction. Every technology adoption requires a change management and digital transformation component, as humans still make the investment decisions.The NLP Problem and the Path Forward
A deeper issue in the world of AI is the lack of standardization around Natural Language Processing (NLP) libraries.
-
NLP is the cognitive part of AI that reads prompts and acts on verbs within the requested context.
-
Current models fail at complex requests (like asking Siri for the average home price over 12 years) because they are programmed for simple searches, not to organize, calculate, and present information based on context.
-
Different large language models are currently creating their own standards, which is like saying, “We’re not going to use Webster’s dictionary ever again”.
-
The long-term solution involves building a blockchain standard for the community, defining what contextually means what, to accelerate AI and improve cognitive capabilities.
Final Recommendations
For professionals and executives looking to succeed in this accelerating age:
-
Ignore the Hype: Ignore ads promoting dozens of tools; they are often the same five or six tools solving only one or two problems.
-
Define the Problem: Before looking for a tool, define the problem you are trying to solve as simply as possible.
-
Hire the Right Experts: Hire a product owner or a digital process engineer to help champion complex adoptions, provide strategic and tactical bullet points, and manage the transition successfully.
-
Embrace Discomfort: Executives need to “get comfortable with being uncomfortable” and find value metrics by identifying their processes and problems.
Watch full episode here:https://dub.sh/E56KpFE
