Liberation Day 2050: Three Possible Agentic AI Futures
With Trump’s tariffs raising the risks of an all-out trade war, it’s suddenly vital for supply chain leaders to look well into the future to develop a resilient plan. One CEO we spoke to last week said the near future is “all about the data,” and how quickly companies can shed those assets that won’t survive the shift, while betting on those that will.
As the shake-up accelerates, digital competence looks like the critical differentiator between winners and losers. And while agentic AI may prove to be the most transformative digital capability of all, its vectors of development could easily wreak havoc without the right values baked in from the beginning. That beginning was last Wednesday.
What Is Agentic AI?
Agentic AI is a software technology that can understand context, make decisions, and execute actions without human oversight. It learns from experience and can operate autonomously as an island of human-like work, or networked with other agents as part of a system able to run on its own.
A robot is a natural vessel for an AI agent, and a swarm of robots (or digital robotic processes) is the same for true agentic AI. This is both the good news and the bad news.
Three Scenarios for 2050
Historians and economists are speculating on the end of globalization as we’ve known it since WWII. Few, however, have asked specifically about the productivity-enhancing potential of agentic AI. Here are three extreme scenarios to consider as plans are made to reconfigure global manufacturing systems.
Optimus Minimus – AI has been called a “species-level threat” by none other than the Godfather of AI, Geoffrey Hinton, formerly of Google. Its capacity to exceed human intelligence has been debated for decades and turned into science fiction images in movies like The Matrix. This story portrays AI deciding that humans are the problem, with Agent Smith declaring us “a virus” and Morpheus lamenting our diminution to batteries powering the machine world. Could agentic AI, released into our factories and logistics networks, decide the world would be better off minimizing humanity? Who can be certain that such an Optimus Minimus scenario isn’t the AI endgame by 2050?
Recommended by LinkedIn
Optimus Rex – AI is already in use militarily, trained to identify threats, tactically engage enemies, and win conflicts. The hostility swirling among nations isn’t yet a hot war, but it is confrontational and based on a zero-sum mentality. The introduction of agentic AI into such a combative system could push economic conflict beyond tariffs, initiating a “king of the hill” dynamic like the Chinese Warring States period (475-221 BC) or the post-Rome Dark Ages in Europe. This Optimus Rex scenario could mean a revolving seat of power occupied briefly by a succession of warlords still raging by 2050.
Optimus Maximus – A recent book called Abundance criticizes the US Democratic Party for losing working-class political support, a shift that’s enabled the current fix we face with tariffs. The authors explore the potential for technology – including AI, robotics, renewable energy, desalination, and vertical farming – to create material abundance greater than ever imagined in history. Agentic AI could connect these technologies in a flywheel of accelerated productivity, maximizing wellbeing for all humans. The Optimus Maximus scenario is a benign master of our physical world trained to feed, clothe, house, and care for everyone by midcentury.
The Role Supply Leaders Can Play Right Now
Agentic AI is still very new. Zero100 data says that only 5% of companies are currently using it in operations. And yet, it’s success in pioneering deployments like SHEIN’s supply chain mesh or Amazon’s SCOT system shows such breakthrough potential that we predict 20% will be using it by the end of this year.
Three essentials will help steer the adoption of agentic AI for the greater good:
1. Declare Your Tenets – Think of these as design principles. They will help with investment prioritization and allow teams to respect your values as they design and roll out new supply chain network assets. These should include respect for safety, community impact, and environmental fallout as well as shareholder returns.
2. Obsess Over Architecture – AI is all about data. Agentic AI is also about control. Design layers of human oversight from the ground up. Prioritize explainability to ensure people understand what is allowed to happen automatically and why. Look to emerging technology tools like guardian agents to automate human oversight where possible.
3. Start With Systems Thinking – Supply chain networks are physical systems tied directly to human health and happiness. Engineering at the systems level necessarily bridges chemical, mechanical, and environmental science as well as finance. Take the extra step to analyze second- and third-order effects of every design decision before casting the die.
The future starts now.
Tech Lead | Supply Chain | "A problem well stated is a problem half solved.” – John Dewey | #WeAreInfios
3wThank you for bringing a great perspective to this topic! AI is not some magic solve-all that we can throw at every challenge and hope for the best. It is simply another tool, granted a very powerful one, that will be hugely beneficial for certain types of tasks. As these systems become more versatile and powerful, their range of responsibilities will grow as well. Agentic AI is certainly the next big step in their evolution. It is therefore our job to manage and direct these tools to obtain the best outcome possible, while minimizing any internal or external risk. Our supply chains link many systems together worldwide. Any unintended results, can thus have far reaching impact. Like a table saw, we need to use our tool where appropriate, feed it with quality data (wood) and ensure we have proper safety measures in place, in case something unexpected happens. After-all, we don't want to be another statistic of someone blindly following their GPS until they drive into a river.