Any views or opinions expressed in this column are of the individual, group, or entity who submitted it and not the publication, owner(s), its advertisers, or affiliates.
Whenever you hear that a corporation is investing in a state, for anything [now] related to AI: data centers, energy infrastructure or others, the next question to ask is that since AI is bound for advancement with all these efforts, if it displaces jobs, what is the readiness of this host state, to protect it’s residents, and what can the state do for others, in a similar service, to return earnings to the state?
By David Stephen
There is a recent report in Franklin County Free Press, Amazon to Invest $20 Billion in Pennsylvania Data Centers, stating that, “Amazon has announced a $20 billion investment in Pennsylvania to build data center campuses in Luzerne and Bucks counties. The project marks the largest economic development initiative in the state’s history and is expected to create thousands of jobs. Senator Dave McCormick (R-PA) praised the announcement, crediting collaboration between Amazon, Governor Josh Shapiro, and various government leaders for securing the deal. The project will also bring workforce training programs to the state. The investment aligns with McCormick’s upcoming Pennsylvania Energy and Innovation Summit, scheduled for next month in Pittsburgh. The event will gather energy and AI industry leaders, government officials, investors, and labor representatives to promote data center growth and job creation in Pennsylvania.”
The State of Pennsylvania must urgently consider setting up an AI Economics Research Lab, as a parallel effort to prepare for job losses—if it happens—indirectly or directly in the state, as a result of AI’s professional proficiency.
This effort would focus on models that can be adopted and then exported, so that PA does not just provide the solution within, but can also lead in that capacity—nationally and globally. In principle, PA would be working on global labor economic models, such that while the preliminary focus would be what can be done within the state, there would be solutions for elsewhere that the lab could explore, as it applies to them, so that PA can get earnings from those efforts.
Although PA needs an AI safety research lab, given the staggering amounts of misuses of AI that is steadily in the news, causing financial, emotional, material and other forms of hurt, AI economics research lab is a path towards solutions that may not require huge compute or data—at least initially, or dependently.
AI Economics Research Lab
The lab has to explore new economic architecture, across roles, rankings, skills, sectors, locations and so forth.
This means that what happens if some accountant loses a job? Or an entry level administrative officer? Or someone that is competent in a software? Or someone in the mining industry? Or someone in some county?
There would be explorations of alternatives, both for remuneration, working hours, new skills and so forth.
For example, would it be possible that instead of unemployment, an individual can work in a job, for lower than the minimum wage but for capped hours?
So, rather than maximum working hours, at full wage, it is possible, for the purpose of keeping people hired, that minimum wage is cut, so long the individual can work for say just 2 hours at a stretch.
While there are roles already, where some people work shifts at full wage and diminutive hours, there are several other roles that would keep people hired, as long as they are working for low hours.
The advantage of this, for example, could be to bring jobs back, reskill people in some key areas, like semiconductors design, fabrication, and so forth.
This will also be adapted in a way that people are hired on different tiers. So, work [below minimum wage] for a while at the lowest tier, before moving up after some weeks. This is better than nothing. Free hours could be [say] spent at the library, to excavate knowledge and note, for a new repository [with slight compensation]. This is a labor experiment that may apply in some form presently but be quite important when AI really kicks in.
While some parts of it may not be applicable in PA, it could be bounded in a software that would properly structure it and then sell to the rest of the world, generating huge incomes to the state. Even if the state cannot do it alone, public-private partnership is an option.
As money becomes more scarce for those immediately displaced from jobs, there would be tasks identification in a county that some people could do, in what can be called a service tax. This means that payment of taxes can be in services, which would be things that should be done that no one would, but required for society. For others, there can be products tax, where they can pay, with what they have, becoming a vault of useful products by authorities.
During the lockdown, there were several things that ran out, without the government having a store of necessities. It could be possible to design a system to pay taxes with products or services for certain individuals or businesses in some situations, becoming a way to ease the pressure on them, have a store, as well as to ensure some form of treasure continuity, regardless of economic tremors.
This store can also be useful to service certain other people, as they can provide some of what they have, towards what they might want. Software too can underscore this.
There would be several models for small businesses, especially on how they can improve. For example, how does a small business grow? What guidance is possible to shape expansion for them, identify rare opportunities and be able to hire more locals?
Texts and email messages can be used to consistently provide them business strategy and intelligence, to ensure that the business is tuned toward vision. For those that engage, there can be customization and closer tutoring. There could be labor capital, to pay wages directly to staff at payment intervals without giving to the owner or manager in bulk. This can be used in seasons to boost sales and marketing.
There can also be new possibilities with language learning for individuals to explore opportunities [remotely] in other jurisdictions—with scarcity. There are some plumbing or other competences in the United States that could be useful elsewhere, and would be willing to pay, so long the American can at least communicate [at a basic level, in the local language]. This may depend on if there is a connector to the marketplace, a payment method, especially at premium. If this does not apply to Americans, it may work for others, but the PA AI economics research lab would provide the structure in a software. Large language models [LLMs] can be used to pervade new marketplaces and AI can determine how news is used to learn a new language.
There are several possibilities, some viable, some complex but everything will be explored, shaped, prepared and then used as paths if AI takes jobs.
The Advantage for Pennsylvania
While AI is the undertone for this project, the actual direction is to create jobs in Penn. and generate massive revenues for the state.
There is nothing new that the AI economics research lab would do that would either not be adoptable in-state or [hyper-valuable] sellable out-of-state or abroad.
So, what is not applicable will become an outlook for revenues, either in the near term or in the future—from elsewhere.
This lab is for new workable [and flexible] models, updating constantly as news emerges of job losses, looking out for how to answer and how answers can be pivotal. This is not going to be a lab of status quo or a lab for the usual talk of Universal Basic Income [UBI] as a way to cater to people if AI displaces jobs.
Some people say AI will not take jobs. Others say AI would. The evidence for now is that there is massive amounts of investments in AI. This could indicate that AI would improve and get better at some of the things it can already do—that humans do for work, and are educated for.
So, whether it can replace people or not is not what is at question, AI is improving, it is best to prepare.
AI Economics Lab Structure
The lab could be situated within a University. The lab will have non-faculty members. The lab may standalone or operate as a public-private partnership, such that it is hosted within a department in one of the organizations, local to PA.
Funding can be minimal at best for this lab, since the core effort is to explore new, flexible, workable and profitable models as economic options in preparation for AI’s purported job displacement.
The lab may also be fully remote, with a few initial members, while expansion is expected and invitations are made.
There may not be a standard or constant remuneration, for members at least at the start of the lab. Questions could be asked on what a member needs and that could be provided. But the work, while not voluntary will not initially bear a standard income.
The direction is to work on what would be so fascinating that it could result in general compensation from the intellectual property at some point or some other form.
While the number of members—for a while—of the lab will be capped, anyone who has anything interesting to offer along the mission will be welcomed to contribute.
June, 2025
The lab can start immediately, if anyone, anywhere in PA can provide minor support. There would be no need to wait until everything is ready before doing so. Immediately the lab starts, some of the initial proposals can be expanded, and then answers can start to be provided, given the news, in AI job displacement at present and what can be done in-state as well as what can be exported.
There is no need to wait. There is nothing to wait for. AI cannot be getting so much investments and not expect that in some form, it would improve in ways that may disadvantage human labor economy.
This could be one of the most important legacy of Governor Shapiro, or of anyone else who steps up.
