Cost-and-Upside Alignment Rule
Make beneficiaries bear their costs or share their upside
- Difficulty
- Moderate
- Time to result
- ~weeks to results
- Steps
- 5
- Confidence
- 91%
Cenk Uygur applies one decision rule across several policy settings: identify who profits, identify who bears the cost, and align the two. A data-center operator should pay for the power its facility requires rather than raise costs for surrounding users. If taxpayers rescue or fund a company, the public should receive equity or another claim on future value instead of absorbing only the downside. If AI businesses create large transition costs through unemployment, he argues that some of their gains should support unemployment insurance or other mitigation. The rule does not settle the disputed size of any future harm. It provides a reusable way to test whether an arrangement privatizes rewards while socializing costs, then redesigns the arrangement so risk, responsibility, and upside travel together.
Origin
In The Diary of a CEO debate, Cenk Uygur reused this logic for data-center power, the 2008 bank rescue, publicly supported drug research, and possible AI-related unemployment. The framework name is descriptive; Uygur did not give the rule a formal title.
Core principles
- 01A private venture should not quietly transfer its operating costs to the public
- 02Public risk or funding should come with a share of the resulting upside
- 03The party creating a measurable social cost should help fund the response
- 04Capitalism works better when costs and rewards remain aligned
How to run it
- 1
Map the beneficiaries
Identify the companies, investors, customers, and public bodies that expect to gain from the activity. Separate direct financial upside from broader claimed benefits.
Pro tip Start with the parties that control the decision and capture the largest measurable return.
- 2
Trace the shifted costs
List infrastructure, fiscal, employment, and community costs that fall outside the beneficiary's accounts. Distinguish demonstrated costs from forecasts or disputed claims.
Watch out Do not treat a predicted cost as established merely because it is politically persuasive.
- 3
Assign responsibility
Require the beneficiary to pay avoidable operating costs and fund mitigation for harms it materially causes.
Pro tip Tie obligations to observable triggers rather than broad rhetoric.
- 4
Price public support
When public funding or risk is unavoidable, negotiate equity, repayment, revenue sharing, or another defined return for the public.
Watch out A public contribution with no reciprocal claim can preserve privatized upside.
- 5
Audit the final balance
Compare who bears downside with who receives upside after the proposed intervention. Revise the arrangement if one side still captures rewards while exporting material costs.
Pro tip Publish the allocation clearly enough that affected stakeholders can challenge it.
In the wild
Uygur argues that a data-center business should pay the full cost of the energy infrastructure it needs instead of raising local users' bills. Kevin O'Leary agrees that a new facility should bring its own generation and says his Utah proposal would also return power to the grid.
→ The project remains possible while its energy burden is assigned to the business expecting to profit from it.
Discussing the 2008 bank rescue, Uygur says the government should have taken equity and later sold it rather than absorbing risk without participating in recovery. He applies the same principle to publicly funded corporate research.
→ Taxpayers receive a defined route to upside when public resources protect or create private value.
Uygur proposes that if AI companies contribute to substantial unemployment, some future gains could fund unemployment insurance. The episode disputes whether that unemployment will occur, so the application depends on verified displacement rather than treating the forecast as fact.
→ Transition support activates when a measurable cost appears and is partly funded by the beneficiaries associated with it.
Common mistakes
Counting only direct expenses
A narrow company ledger can omit infrastructure or transition costs borne by communities and taxpayers.
Taking risk without a return
Public support can leave taxpayers carrying downside while private owners retain all future gains.
Treating forecasts as facts
The rule is strongest when obligations respond to measured costs rather than unverified predictions about future harm.
Is it for you?
Best for
It is best for evaluating projects, subsidies, bailouts, and technologies that create both concentrated gains and widely distributed costs.
Not ideal for
It is not ideal when costs, beneficiaries, or causal responsibility cannot be estimated with enough confidence to assign them fairly.
From the transcript
“if you're going to do a data center you have to pay for it and you have to pay for all of its costs”
“if we're going to pay your costs, obviously we should have equity in it”
“They created this cost which is unemployment. Why don't they pay for it?”
From the episode
EMERGENCY DEBATE: The Economy Is About To Collapse! The 2026 AI Crisis Nobody Sees Coming