The Big Picture · July 2026
Insurers buy the expert opinions that let them deny claims they owe, then game the legal system to bury the proof so the scheme keeps running.
You buy insurance to transfer a risk you cannot afford to carry alone: the house, the disability, the medical bill that would otherwise end you. You pay for years against a day you hope never comes. Then the day comes, you file the claim you paid for, and a company you have never met hands your file to a physician or an engineer it has paid hundreds of times before. And that expert finds, as the expert nearly always finds, that what happened to you is not covered, not as bad as you say, or not real.
That is the transaction at the center of everything this publication studies. It is a method, and it runs at scale: the systematic use of financially dependent experts to manufacture the appearance of a legitimate dispute, so that a claim the company owes can be denied or underpaid with a defensible paper trail. Call it what it is. It is a way of cheating claimants out of money they are owed, and by its scale and durability it is one of the largest institutionalized frauds in this country's history. It has run more than forty years, and only in roughly the last fifteen have courts confronted it in earnest.
An insurer that wants to deny a claim it owes cannot simply say no; a flat refusal is bad faith, and bad faith is expensive. What it needs is a reason, an opinion from someone with a credential that the claim fails. So it retains an expert. And the expert it retains is not neutral, because neutrality is not what the arrangement selects for. The reviewer whose income, year after year, comes from the companies whose claims they evaluate learns quickly which findings renew the engagement. Nobody has to instruct them. The incentive does the work. The opinions that result have a recognizable shape: authoritative-sounding and untestable, applying standards found in no policy (the notorious demand for "objective evidence" of conditions that have none), reviewing paper without examining a person, all pointed the same direction.
Insurance is not an ordinary contract. A policy hands the company a discretionary power over something you cannot replace, the decision whether to pay when you are at your most vulnerable, and the law responds by imposing a duty ordinary contracts do not carry: the covenant of good faith and fair dealing, which in nearly every state requires the insurer to exercise that discretion reasonably and honestly, and in California requires a "thorough, fair, and objective investigation" by regulation. An investigation skewed by a financially dependent expert is the precise opposite of that. The breach is not incidental to the practice; the practice is the breach. This is why insurance is treated differently, and why a denied claim, on these facts, is a breach of a heightened duty rather than a disagreement.
Most claimants never reach that framework, because they fight the wrong battle. Handed a denial backed by an expert opinion, they answer with an expert of their own and argue the merits: mine is better qualified, investigated more carefully, and reached the right result; theirs got it wrong. That is the battle of the experts, and it is a losing proposition. It asks a jury to choose between two credentialed opinions, and even when the claimant's expert prevails, the insurer has its escape ready: it relied on a qualified expert, so it cannot have acted in bad faith, and the punitive damages fall away. Fighting the merits feeds that defense; attacking the bias takes it away.
A duty you cannot prove was breached is a duty in name only. The correction is an operational framework for proving bias claim by claim, reduced to four factors a practitioner can run against a live claim file:
Around these sits a standard, inference of bias, a century-old rule that a decision-maker with a financial stake is suspect by the structure of the arrangement, and a presumption: once the claimant makes a modest showing on the checkbook, the burden shifts to the insurer to show its expert was actually neutral, and silence where records could have been kept is not treated as neutral.
Every one of those factors depends on evidence that exists only in the insurer's, the vendor's, and the expert's files. An insurer that successfully resists that discovery has, as a practical matter, defeated the bias claim regardless of its merits. So the second half of the scheme is procedural: when a claimant asks for the proof, the insurer games the discovery rules to keep it buried. The playbook is to raise every objection on every request (irrelevant, overbroad, unduly burdensome, a "fishing expedition," an invasion of privacy, a "mini-trial") not because any one is strong, but because the aggregate raises the cost of proof until some claimants stop. The Discovery Objections Series takes that playbook apart objection by objection: each has a codified answer, the labels name proceedings that do not exist, and a claimant who drafts around the objections turns the insurer's own obstruction record into evidence for the bad-faith case.
The claim-by-claim work would matter less if a public regulator were policing the practice. None is. The record comes from this project's own primary analysis: California's insurance regulator has achieved zero adjudications for biased-expert use in its sixty-seven-year enforcement history, and across more than fifty years and fifty-one jurisdictions, no insurance commissioner anywhere has adjudicated a biased-expert case. Private bad-faith litigation is not one enforcement mechanism among several; in practice it is the only one that works, which is exactly why the evidence that makes it work is under political attack.
That is the through-line into the legislative fight this publication has documented in real time. In California, a trio of bills advanced under a consumer-protection banner while doing the opposite. SB 354, styled a privacy law, in its original form would have authorized systematic destruction of the claim records and AI decision systems that wildfire survivors and other claimants need to prove institutional bad faith; sustained opposition closed the front door (the deletion mandate) and then the back door (an Administrative Procedure Act exemption that would have let the retention standard be set by unreviewable order). AB 1680 would route those same survivors into a surplus-lines market outside the regulator's reach as their records thin. AB 1795 seated an industry-aligned panel, with no toxicologist, to write the smoke-damage standards those survivors would be measured against. Together the three complete a program: clear the evidence, remove the policyholder from oversight, capture the standard.
Bias in the claim file and evidence-destruction in the statehouse are the retail and wholesale versions of one scheme.
Insurance is the instrument a society uses to make catastrophe survivable, to let people rebuild after a fire, keep a home after a disability, get treatment after a diagnosis. A claims system that quietly bends its own experts toward denial does not just wrong the individual claimant. It degrades the thing insurance is for, for everyone who holds a policy and has not yet had to test it. And what it takes is not abstract: money and property outright, the claim owed and never paid; liberty, the years of litigation, debt, and displacement forced on people who did nothing but file a claim they were entitled to; and in the starkest cases, where a denial cuts off disability income or medical treatment on a manufactured opinion, health and sometimes life itself.
The reason to be hopeful is the same reason the apparatus is worth mapping. Nothing in the correction requires a new cause of action, a new regulator, or a new theory of law. The duty is settled. The standard is a century old. The factors are already discoverable and the discovery is already winnable. The presumption is already recognized. What the practice has counted on for forty years is not that the law is on its side. It is that too few people are positioned, or willing, to name it plainly and prove it, claim by claim, statute by statute. That is the work.
Read the essay
The full essay on Expert Bias Report, tying the scheme, the doctrine, the practical tools, and the legislative and regulatory fight into a single map, and the case that the apparatus can be dismantled.
The Discovery Objections Series → The bias-evaluation service → The legislative record →
Subscribe to Expert Bias Report →
This essay synthesizes work published on Expert Bias Report between June and July 2026: the four-factor framework, the Discovery Objections Series, and the legislative correspondence on SB 354, AB 1680, and AB 1795, with the project's primary analysis of the regulatory enforcement record and the pattern-evidence discovery corpus. Empirical claims about the enforcement record and the discovery corpus derive from the project's own analysis. Educational and informational only; not legal advice.