AI

Political record and interview follow-up

Abdul El-Sayed connects this existing policy analysis to the running tracker. El-Sayed - AE-003 - AI Oversight and Research Pause separately examines the September 2026 interview proposal and its unresolved institutional details.

AI, Automation, and Blockchain Policy Proposals

The proposal argues that powerful AI and blockchain should not be governed mainly by private corporations. It supports federal guardrails for AI deployment, labor protections against automation, safety rules for frontier models, compute-provider oversight, and an independent AI agency with enforcement power. On blockchain, it favors legal clarity for legitimate financial uses while cracking down on fraud, laundering, pump-and-dumps, and political corruption.

1. AI Should Be Regulated Before It Is Widely Deployed

Proposal:
Create legal guardrails around when, how, and for what purposes AI is used. Oppose federal pre-emption of state AI laws unless Congress first passes strong federal legislation.

Best argument for it:
AI systems can affect jobs, elections, privacy, education, policing, finance, cybersecurity, and national security. If government waits until after harms occur, regulation may arrive too late. NIST already has an AI Risk Management Framework, but it is voluntary, so this proposal pushes toward enforceable standards. Source: NIST AI Risk Management Framework

Counterarguments:

• Overregulation could slow innovation and make U.S. companies less competitive against China and other rivals.
• State-by-state AI rules could create a confusing compliance maze for startups.
• Regulators may not understand the technology well enough to write rules that stay useful.
• “AI safety” language can be used by large companies to lock in their dominance by raising compliance costs.

2. AI Should Enhance Human Life, Not Replace Workers

Proposal:
Address mass automation through wage insurance, basic income, union investment, tax reform that stops favoring machines over workers, and small-business revolving loan funds.

Best argument for it:
If AI increases productivity but concentrates the gains among firms and shareholders, workers may face job loss, wage pressure, or bargaining-power collapse. Wage insurance and UBI-style policies could soften the transition.

Counterarguments:

• Automation may create new jobs rather than simply destroy old ones, as previous technologies did.
• Basic income could be extremely expensive and politically unstable.
• Taxing automation may discourage productivity growth and make goods more expensive.
• Wage insurance can preserve declining industries instead of helping workers move into better ones.
• Stronger unions may help some workers but could also reduce flexibility in fast-changing industries.

3. Frontier AI Needs Security and Safety Rules

Proposal:
Require safety testing for frontier models, liability rules, incident reporting, compute controls, “know your customer” rules for compute providers, and an independent technical agency similar to the FDA.

Best argument for it:
The most powerful models may create risks in cyberattacks, biological misuse, fraud, autonomous agents, and critical infrastructure. Mandatory testing and incident reporting would make AI development more accountable. NIST’s generative AI profile already identifies unique risks from generative AI, though it remains voluntary. Source: NIST Generative AI Profile

Counterarguments:

• A new AI agency could become slow, politicized, or captured by the biggest AI companies.
• FDA-style approval may be too rigid for software, where models update constantly.
• Compute controls may push development offshore or into black markets.
• “Know your customer” requirements for compute could create surveillance concerns.
• Liability rules may be hard to apply because AI harms can involve model developers, deployers, users, and downstream integrations.

4. Blockchain Should Be Legal but Regulated

Proposal:
Clarify legitimate blockchain uses in financial services while protecting consumers from fraud, money laundering, pump-and-dump schemes, scam coins, and corruption.

Best argument for it:
Blockchain and stablecoins may improve payments, settlement, and access to financial services, but crypto markets have also produced scams, opaque reserves, money laundering risks, and political conflicts of interest. Regulation could separate useful infrastructure from predatory speculation.

Counterarguments:

• Heavy regulation could kill useful decentralized innovation before it matures.
• Existing fraud, securities, banking, and anti-money-laundering laws may already cover many abuses.
• Regulators may treat all crypto as suspicious instead of distinguishing legitimate projects from scams.
• Consumer-protection rules can become a backdoor way to protect banks from competition.
• Blockchain’s promise of decentralization weakens if every project must fit traditional financial regulation.

Bottom Line

The proposal is broadly progressive-populist and technocratic: it distrusts concentrated corporate power, wants government to shape markets before harms become irreversible, and favors worker protection plus public-interest oversight. The strongest critique is that it may overestimate government competence, underestimate innovation costs, and accidentally entrench the largest firms it seeks to restrain.

Abdul El-Sayed: Income Protection and Worker Power Against AI Automation

El-Sayed’s proposal treats AI automation as both an economic shock and a power problem. His answer is not only to regulate AI, but to protect workers from displacement through wage insurance, possible basic income, stronger unions, tax changes that stop favoring automation over human labor, and public investment in small businesses.

1. Wage Insurance

What it means:
Wage insurance would partially replace lost earnings when a worker loses a job and has to take a lower-paying one. Unlike unemployment insurance, it does not only pay people while unemployed; it helps after reemployment if the new job pays less.

Why El-Sayed would support it:
This fits an AI-displacement world where workers may not become permanently jobless, but may be pushed from stable, middle-income work into lower-paid service or gig work. Wage insurance tries to preserve income during that transition.

Best argument for it:
It rewards people for returning to work while cushioning the income shock from automation. It is more targeted than universal basic income because it helps people who can prove actual wage loss.

Counterarguments:

• It may subsidize employers who offer lower wages, because government fills part of the gap.
• It helps displaced workers after harm occurs but does not prevent job loss.
• It may be hard to administer fairly across gig work, freelance work, and irregular income.
• It may leave out people who cannot find new work at all.
• It assumes the main AI harm is lower wages, not permanent exclusion from the labor market.

2. Basic Income

What it means:
Basic income usually means regular cash payments, often unconditional, meant to provide a minimum income floor. In the AI context, it is often framed as an “AI dividend”: if machines produce more wealth, citizens should share in the gains.

Why El-Sayed would support it:
If AI creates mass productivity while reducing the need for human labor, then work-based welfare may become less adequate. Basic income would preserve purchasing power and give people more freedom to retrain, care for family, start businesses, or refuse exploitative work.

Best argument for it:
It addresses the deeper possibility that AI does not merely cause temporary job transitions but weakens the connection between work and survival. It also gives workers bargaining power because they are less desperate.

Counterarguments:

• A meaningful national UBI would be extremely expensive.
• If funded poorly, it could require large taxes, cuts to existing welfare programs, or deficit spending.
• If payments are too low, it becomes symbolic rather than transformative.
• If payments are high, critics argue it could reduce labor-force participation.
• Some labor advocates worry UBI could become a substitute for good jobs, public services, and union power.
• It may not solve housing, healthcare, childcare, or education costs if those markets remain expensive.

3. Worker Investment Through Unions

What it means:
This means strengthening unions so workers have a collective voice in how AI is introduced, how productivity gains are shared, and whether displaced workers receive retraining, severance, wage guarantees, or job protections.

Why El-Sayed would support it:
El-Sayed is strongly pro-union and has linked union power to worker dignity in an AI economy. Reporting on his campaign notes that he argues workers deserve the right to unionize and that AI should be regulated to protect people from losing work to automated systems. Source: Bridge Michigan

Best argument for it:
Unions can bargain over the actual workplace deployment of AI, not just clean up after the damage. They can demand notice, transparency, retraining, profit-sharing, limits on surveillance, and protections against algorithmic firing.

Counterarguments:

• Unions can protect existing members while making it harder for outsiders, younger workers, or small firms to compete.
• Stronger unions may slow adoption of productivity-enhancing technology.
• Union rules can become rigid in industries that need fast adaptation.
• Unionization works better in stable workplaces than in gig, freelance, or highly fragmented sectors.
• If AI changes entire industries, firm-level bargaining may be too narrow to solve the broader problem.

Overall Assessment

El-Sayed’s package is best understood as a worker-power response to AI, not just a welfare proposal. Wage insurance handles downward mobility, basic income handles broader insecurity, and unions try to give workers leverage before automation decisions are made.

The strongest version of the proposal is a mixed system: wage insurance for displaced workers, unions for workplace bargaining, and targeted cash support if AI creates broad labor-market instability.

The strongest critique is that the plan is not yet fully specified. The big unanswered questions are: How large would the benefits be? Who qualifies? How is it funded? Would it slow innovation? And how do these policies avoid helping large firms comply while crushing smaller competitors?

Abdul El-Sayed on AI Data Centers

Abdul El-Sayed’s data-center position is that Michigan should not let AI firms and utilities dictate terms to local communities. He argues that data centers may strain electricity, water, grid reliability, local democracy, and household utility bills. His “terms of engagement” demand no rate hikes, community transparency, reliability guarantees, job guarantees, closed-loop water systems, enforceable community benefits, no clean-energy loopholes, and real penalties for violations.

Core Claim

El-Sayed frames AI data centers as a public-utility and community-control issue, not just a tech-development issue. His concern is that large AI companies and utilities can privatize profits while shifting infrastructure costs, environmental burdens, and grid risks onto ordinary residents.

Michigan reporting says his campaign highlighted more than 15 proposed data-center projects in the state, including a major Oracle/OpenAI-linked project in Saline Township. Source: Michigan Advance

1. No Utility Rate Hikes

Proposal:
Data centers should pay for their own energy demand. Costs should not be passed onto regular ratepayers.

Why he supports it:
AI data centers require huge electricity loads. If utilities build new infrastructure to serve them, El-Sayed worries households could end up paying through higher rates.

Strong argument for it:
This prevents “corporate welfare through the electric bill.” If a private tech firm creates the demand, it should pay the full cost.

Counterarguments:

• Michigan regulators already claim they are adding protections so data centers do not shift costs to other customers. Source: Michigan Public Service Commission
• Large-load customers can sometimes lower rates for everyone if they contribute heavily to fixed grid costs.
• “No rate hikes” may be too absolute because utility rates are affected by many factors besides data centers.
• If rules are too strict, projects may move to states with looser requirements.

2. Community Transparency and Local Say

Proposal:
Communities should know the expected impacts on electricity rates, water, reliability, jobs, and infrastructure before projects are approved.

Why he supports it:
He argues local communities are being outmatched by tech companies, utilities, and state-level regulators.

Strong argument for it:
Data centers can reshape small communities. Residents deserve disclosure and participation before long-term utility and land-use commitments are locked in.

Counterarguments:

• Too much local veto power can create endless delays for infrastructure that has regional or national economic value.
• Some contract details may involve trade secrets or security-sensitive information.
• Local opposition can become NIMBYism even when projects bring jobs, tax revenue, and grid investment.
• State-level regulation may be more consistent than township-by-township decision-making.

3. Grid Reliability Guarantees

Proposal:
Data centers should not weaken electric reliability for surrounding residents.

Why he supports it:
A large AI campus can use electricity at the scale of a city or power plant. El-Sayed worries that utilities may prioritize data-center demand while ordinary customers face outages or degraded service.

Strong argument for it:
If a data center draws massive power, it should be required to fund grid upgrades, backup systems, and reliability protections.

Counterarguments:

• Data centers can actually help finance grid upgrades that benefit everyone.
• Long-term contracts can require minimum payments and reduce stranded-cost risk.
• The real problem may be weak utility regulation generally, not data centers specifically.
• Reliability rules already exist through utility commissions and grid operators.

4. Water Protections and Closed-Loop Systems

Engineering and evidence bridge

AI Data Centers and Water Use shows why “closed loop” is an incomplete regulatory standard: the server coolant can recirculate while a separate cooling tower or adiabatic stage still evaporates water. A stronger rule specifies non-evaporative heat rejection or enforceable limits for WUE, monthly consumption, source-water mix, dry-season withdrawals, and public operating data.

Proposal:
Data centers should use closed-loop water systems and avoid stressing or polluting local water resources.

Why he supports it:
AI data centers can use large amounts of water directly for cooling and indirectly through electricity generation.

Strong argument for it:
Water is a local resource. Communities should not be forced to trade water security for private AI infrastructure.

Counterarguments:

• Newer data centers can use less water through air cooling, liquid cooling, or recirculating systems.
• Water impact depends heavily on design, location, and energy source.
• Strict water rules may push development to places with weaker environmental oversight.
• Energy consumption may be the larger issue than direct water use in some regions.

5. Job Guarantees and Community Benefits

Proposal:
Data-center developers should provide binding community benefits: jobs, grid upgrades, water treatment, piping, and other local infrastructure.

Why he supports it:
El-Sayed wants communities to receive concrete benefits rather than vague promises.

Strong argument for it:
If a data center consumes public resources and reshapes local infrastructure, the community should receive enforceable economic gains.

Counterarguments:

• Data centers often create many construction jobs but relatively few permanent jobs.
• Forced community-benefit agreements can become political bargaining tools.
• Too many required benefits can make projects economically unattractive.
• Benefits may be promised broadly but captured by politically connected groups.

6. No Clean-Energy Loopholes

Proposal:
Data centers should not be allowed to sidestep clean-energy laws or increase fossil-fuel dependence while claiming to support innovation.

Why he supports it:
AI infrastructure could dramatically increase electricity demand, making climate goals harder to meet.

Strong argument for it:
If AI companies claim to be building the future, they should not do it by increasing dirty-energy demand or weakening state climate laws.

Counterarguments:

• Data centers can accelerate renewable investment by creating large, stable demand.
• Some firms may buy renewable power or fund storage and transmission upgrades.
• Clean-energy mandates may raise costs or slow urgently needed AI infrastructure.
• The emissions question depends on the actual power mix, not simply the existence of a data center.

Overall Assessment

El-Sayed’s data-center position is a populist-regulatory argument: AI infrastructure should serve communities, not overpower them. His strongest point is that data centers create real local externalities: electricity demand, water use, land-use conflict, secrecy, and possible ratepayer risk.

The strongest counterargument is that Michigan already has regulatory tools to protect ratepayers, and data centers may bring investment, construction jobs, utility revenue, and grid upgrades. The real policy question is whether strict front-end rules prevent abuse or simply drive AI infrastructure elsewhere.