Mark Cuban’s Plan to Make Healthcare Affordable

UnitedHealth Group

Across the shows: 2 speakers in other episodes, 2 bullish and 0 bearish calls, past 30 days

• Mark Cuban · Cost Plus DrugsBearish

  UnitedHealth's vertical integration lets it raise premiums and steer members to its own owned units.

  UnitedHealth has more than 3,000 subsidiaries and roughly $150 billion in intercompany transfers a year, which Cuban says would make it the 14th-largest economy among US states. Because of that scale, the insurer can raise premiums, deductibles and out-of-pocket costs while redirecting members into relationships with companies it already owns, often without the employer knowing. Cuban says CVS and Cigna are not far behind. He backs the Break Up Big Medicine bill to break up these companies and open the market to entrepreneurs.

  “So UnitedHealth, 3,000 plus subsidiaries, [$]150,000,000,000 in intercompany transfers annually. ... they will increase premiums, increase deductibles, increase out of pocket costs, and at the same time redirect you or actually force you into a relationship with a company they already own”at 11:25

Healthcare system & insurers

• Julie Yoo · Andreessen Horowitz (a16z)

  Employer health premiums are surging, and startups are now entering the market that was closed to them.

  Employer-sponsored insurance covers more than half of Americans, but the market has been dominated by national carriers with entrenched networks, which made startup entry too costly. Yoo says some carriers are forcing 60% year-over-year premium increases even on her portfolio startups, and that premiums are rising faster than people realize. She argues AI lowers the cost of building a full-stack health plan, which is driving a wave of startups to enter the market and reduce the friction and information asymmetry in benefits contracts.

  “the premium, increases are just freaking ridiculous. This year we're seeing, even amongst our startups, some BUCAS are forcing 60% year over year increases in premiums.”at 11:08

• Mark Cuban · Cost Plus Drugs

  Insurer-owned administrators hide spread pricing, and AI cannot protect employers once contracts are signed.

  Insurer-owned administrators (ASOs) and their large contracts hide the spread between what employers pay and what hospitals are paid, and they make reconciliation nearly impossible. Employers often cannot see their own claims data. Cuban cites the City of Dallas, which had not been able to see its 135-page ASO contract. Cuban says AI can help avoid signing a bad contract before signing, but once signed, the insurer's terms win because no human or AI can catch every term.

  “...But once you sign that contract with terms you don't fully understand and conditions you can't even protect yourself against, they win. There's no AI in the planet that can protect you.”at 15:46

• Mark Cuban · Cost Plus Drugs

  If employers ran benefit contracts through AI before signing, the big insurers and PBMs would collapse.

  Employers covering more than 160 million Americans, about 67% of the market, hold most of the leverage. Cuban argues that if they used AI to audit contracts before signing and negotiated from that, the system behind the big insurers and PBMs and their subsidiaries would collapse, and healthcare costs would drop sharply. He urges companies to appoint a healthcare CEO or CFO whose job is contract management, and says a few public companies that did this were shocked at how much they had been overcharged.

  “If that 67% of the employers representing 67%, the 160,000,000 plus Americans, actually did one thing, ran their contracts through AI before they signed them and negotiated from that, the entire system for the big BUCAS and big PBMs and all their subsidiaries would collapse.”at 21:26

• Mark Cuban · Cost Plus Drugs

  Publishing taxpayer-funded hospital contracts and ledgers would create an efficient market and lower costs.

  Cuban's first policy recommendation is that all contracts with any entity taking taxpayer money be made public with no redactions. He argues that once this happens, an efficient market forms and costs fall. He also wants nonprofit hospitals to publish their full general ledger and procurement receipts, since it is taxpayer money. He adds that people should be able to apply a cheaper price they find to their deductible, which eight states already allow.

  “One, make all contracts with anybody who takes taxpayer money public. No redactions. Because the minute that happens, you start to have an efficient market and the costs go down.”at 48:59

• Mark Cuban · Cost Plus Drugs

  Big carriers take little risk now; 72% of their revenue comes from taxpayers.

  Cuban says 72% of big insurance carriers' revenue comes from taxpayers. He argues they no longer take risk and instead offload it, so everything is an arbitrage for them. He adds that insurance premiums are the one debt people never pay off, since they keep paying until they die, even under Medicare. He concludes that the current insurance model is extractive and that people have to fight for their benefits when they need them.

  “72% of those big carriers' revenue comes from taxpayers. They don't take risk anymore. They [offload] it. Everything's an arbitrage for them.”at 49:51

• Mark Cuban · Cost Plus Drugs

  A deposit-and-loan health account with federal stop-loss would make insurance companies obsolete.

  Cuban proposes that people deposit the equivalent of an ACA silver premium into accounts they own, for example $2,100 a year for a Texas family of five. A bank would lend against the account, a stop-loss would kick in above a threshold such as $40,000, and the federal or state government would guarantee the loans, with repayment deducted from payroll. Anyone who never has a major incident keeps the deposits and accrued interest at age 65. He argues that because the money belongs to the depositor, people would not have to fight insurers for coverage, and that insurers would no longer be needed in this structure.

  “All of a sudden, bye bye insurance companies. Now all of a sudden, you're not fighting back because it's your money to spend as you choose.”at 52:26

AI & jobs

• Mark Cuban · Cost Plus Drugs

  AI does not replace jobs; people who know how to use it replace those who don't.

  Cuban argues that entry-level roles cannot simply be cut, because models are always behind the latest research and people need to stay in the loop, provide context and grow with the organization over time. He says the bigger risk is workers who do not learn to use AI, and that companies need to educate their C-suites on how models, agents and open versus closed models work before they can adopt AI well.

  “AI doesn't replace jobs. People who know how to use AI replace the jobs of people who don't know how to use AI.”at 28:28

• Mark Cuban · Cost Plus Drugs

  AI will not replace doctors, but will create jobs as translators between AI output and patients.

  Cuban says he will never replace doctors and that trying would be a mistake. Latency and the cost of processing, even as compute gets cheap, mean patients still need someone who knows which questions to ask and which variables matter. A human with all five senses can read context, such as a crying family member, that a robot on video cannot. He expects medical school training to change, with doctors learning to use tools like OpenEvidence and to translate a patient's DNA data and the model's output into something the patient understands. He also expects a new job category of AI translators who extract the novel data a patient's model needs and make the AI's output valid for an individual.

  “But I'd rather have a human there with all five senses being able to take a look and ask questions... And so I don't think ever that we'll replace doctors and we shouldn't even try.”at 40:52

• Julie Yoo · Andreessen Horowitz (a16z)

  Cheaper AI care advice will increase demand for physicians, who will shift toward physician-scientists.

  Yoo argues that as care advice becomes more abundant, more people will demand real-life healthcare services, so the country will need more doctors. She says today's doctor role is largely rote, such as checking quality measures, and that personalized drugs could be one of the most disruptive forces for care delivery. In her view, the job could become that of a physician-scientist who collects novel data and generates insights at the edge of knowledge about how individuals respond to therapies. She also expects longitudinal wearable monitoring, such as daily blood pressure, to deepen that individual data.

  “I think we're gonna need more doctors in the future because as care advice becomes more abundant, more people are gonna demand healthcare, actual, like real life healthcare services.”at 43:53

AI: overall outlook

• Mark Cuban · Cost Plus Drugs

  AI cannot fix healthcare while its economics and incentives stay broken; rival AIs will just net out.

  Cuban is a big AI fan, and sees it as important clinically, operationally and most of all short term on the contract side. However, he argues that introducing AI into a broken financial system does not fix it, because there is no price discovery and the contracts are so large and long that insurers keep squeezing. He calls AI-driven claims and billing 'bot wars' short-term and says AI is a great equalizer: every side gets the same tools, so the gains net out and he does not expect dramatic long-term change. He also thinks insurer-owned silos make it hard for them to become AI-native.

  “But the reality is in a broken financial system, even when you introduce AI into the economics and the operations and the processes, When it's already broken, you can't fix it.”at 14:44

GLP-1s & obesity drugs

• Mark Cuban · Cost Plus Drugs

  Food makers are unprepared for GLP-1 disruption; AI can design cheaper GLP-1-friendly rivals in minutes.

  Cuban recounts a meeting with a consumer food company head who was discussing acquisitions because of GLP-1 drugs. Cuban asked ChatGPT to design a cheaper, GLP-1-friendly competitor to the company's top cookie, complete with a full bill of materials and sources. It produced the answer in about three minutes, and the executive realized he had not considered the threat. Cuban uses the example to show that many companies do not understand how AI can be used against their products.

  “let's take your number one product, your cookie, and I'm going to ask ChatGPT to create a competitor that is more GLP-one friendly and will be priced lower... in about three minutes we got back to it, he was like, Oh, shit.”at 30:03

Pharma & biotech

• Mark Cuban · Cost Plus Drugs

  AI models will make personalized drugs cheap, starting with over-the-counter medicines and then prescriptions.

  Cuban argues the cost of designing a drug personalized to a patient's DNA will approach zero as model capabilities rise, and that algorithmic efficiency gains, not just GPUs, will drive this. He gives no fixed timeline, saying it could take five, ten or twenty years. He expects it to start with over-the-counter drugs like Advil, Tylenol and ibuprofen, where everyone currently takes the same dose and must experiment to learn how it affects them. Personalized prescriptions such as choosing between Eliquis, Xarelto and warfarin would follow. He expects initial costs to be enormous but to fall until personalized medicine becomes part of a normal subscription, similar to how DNA testing fell to about $99.

  “The cost of designing a drug personalized to you after reviewing your DNA and knowing everything that it knows will start approaching zero as the capabilities of models go up.”at 36:44

AI pricing & unit economics

• Mark Cuban · Cost Plus Drugs

  AI inference costs are high today but falling very quickly, making consumer AI subscriptions cheaper to serve.

  Cuban says the inference and processing needed to respond to a request is expensive today but coming down very quickly. He expects the same to happen for personalized medicine. From this he argues that model companies' $20 or $200 consumer subscribers should add personalized health services once they are available, since the marginal cost of delivering them will be small.

  “the inference and everything that has to happen just to respond to a request today. Expensive like this, right? Coming down very quickly, but the same thing's going to happen for this type of personalized medicine.”at 38:46

AI regulation & safety

• Mark Cuban · Cost Plus Drugs

  IP owners will charge for model access, complicating AI training and medical innovation.

  Cuban argues that patents will be hard to protect, since competitors can change a patent slightly and file their own. Doctors who create new procedures, he says, will require payment per access or force users onto their software for inference, because nobody gives away valuable IP for free. He calls this the central tension in the AI-entrepreneur view that information wants to be free, and says it will change how people interact with models. He argues the industry has not yet addressed where the IP needed to train, post-train or retrieval-augment a model will come from if owners do not make it available.

  “Nobody's going to give away that IP for free. And that creates a lot more problems.”at 43:03

Big Tech platforms & digital ads

• Mark Cuban · Cost Plus Drugs

  Social media algorithms, shaped by AI, drive political divisiveness; platforms don't care what they feed.

  Cuban argues social media algorithms determine how most people in the country think, and that each person now has a unique feed. He cites a 23-year-old friend whose friends did not know Texas candidates, as evidence that people get more of what they seek. Platforms do not care what they feed users beyond minimal constraints. He sees the irony that the AI behind these algorithms is what ruins things. He says Section 230 keeps this in place unless the US adopts something like Australia's or some states' under-16 restrictions. He also argues frontier AI labs must become part of everyone's feed or else they will keep publishing useless position papers.

  “Because what controls how pretty much everybody in this country thinks right now? Social media algorithms.”at 53:18

AI infrastructure & data centers

• Mark Cuban · Cost Plus Drugs

  Data center backlash is a proxy for poor AI understanding; communities struck bad deals.

  Cuban argues that opposition to data centers is a proxy for people not understanding AI, and he says he does not blame them for that. He also argues that many communities negotiated poor deals for data centers, falling for what he calls contractual terrorism. He treats the backlash as a fixable problem of education and negotiation rather than a reason to slow buildout.

  “Data center hate is just a proxy for not being able to understand AI... a lot of communities have been really dumb with the deals they've negotiated for data centers.”at 55:38