Strategic Acquisition Brief · Confidential
Brian Walsh has described the two halves of good guidance precisely: what the math and the science say, and how you position it so people actually act. Coach is very good at the second half. The first half is still a language model reasoning about money — which is why it ships to 14.7 million members with “responses are information for you to consider and not financial advice.” MaxiFi is the science: computationally exact, economics-based planning — for a household’s facts and assumptions, it solves, not guesses, the lifetime plan, every dollar of taxes and benefits computed under current law. Deterministic, reproducible, auditable. Built over 30 years by BU economist Laurence Kotlikoff.
In June, SoFi acquired Composer Securities and shipped it as Composer by SoFi within weeks — natural language in, a rules-based, backtestable strategy out. Anthony Noto framed it as the strategy plainly: identify innovative technologies and exceptional teams that strengthen the ecosystem, because AI is becoming a foundational part of investing.
Days earlier, SoFi Coach went live to SoFi Plus members: an AI chat that connects across 12,000 financial institutions, built alongside the firm’s own financial planners, with roughly 70% of engaged test members taking a meaningful financial action. The inputs problem — the thing that kills most planning tools — is already solved here.
Coach’s own language is that its responses are information to consider and not financial advice, based on limited information from connected accounts. That is an honest and correct disclosure for a model-generated answer. It is also the ceiling on what “get your money right” can mean.
Composer is the precedent for how this gets fixed: acquire the computation, put the SoFi interface on it, ship it in weeks. The difference is that a lifetime plan is a larger prize than a trading strategy — and a harder computation that nobody else in consumer fintech owns.
MaxiFi is not an app SoFi would operate alongside its own. It is a computation service Coach calls, the same architectural position Composer now occupies behind the investing tab. The member never sees it. What changes is what Coach is able to say.
Coach, the SoFi app, SoFi Plus. Same interface, same conversational surface, same team of planners shaping how answers are positioned.
12,000 institutions connected. The account aggregation that most planning engines never get is already in place, which is why this integrates in weeks rather than quarters.
The rules, the solver, the audit trail. Same inputs, same answer, every time, traceable to the law tables in force on the plan date.
A computed, auditable number is a different object from a generated one. “Information for you to consider” becomes something SoFi can stand behind.
That division is not a compromise; it is the reference architecture for high-stakes AI. Where a wrong answer is catastrophic, no serious operator ships the raw model to the user. Lifetime planning is exactly such a function: the member who retires at 62 on a number that was approximated, the family under-insured by a million dollars.
Walsh had it right — here is what the math and the science say, and here is how humans actually behave. Coach owns the second half already. This is the first.
MaxiFi is the financial-planning platform of Economic Security Planning, Inc., built over more than three decades by Professor Laurence Kotlikoff of Boston University. It uses consumption smoothing and dynamic programming to compute the single, mathematically optimal lifetime plan — solving simultaneously across Social Security strategy, federal and state taxes, Roth-conversion sequencing, withdrawal order, life-insurance need, estate planning, and upside investing.
Goals-based tools and rule-of-thumb calculators answer “What is the chance you hit your number?” MaxiFi answers “What is the optimal path, and how much can I spend today without jeopardizing tomorrow?” It is not a better simulator. It is a different class of engine.
Prof. Laurence Kotlikoff — William Fairfield Warren Professor at Boston University; Harvard Ph.D.; former Senior Economist on the President’s Council of Economic Advisers; named by The Economist among the 25 most influential economists. He intends to stay on with the acquirer in whatever capacity best serves the product — architect, spokesperson, advisor.
MaxiFi’s economics build on Nobel-laureate work, and Nobel laureate Robert Merton teaches with MaxiFi at MIT Sloan as an “outstanding science-based lifecycle and retirement management platform.” Featured in Bankrate’s “Best financial planning software of 2025” roundup, cited as best for near- and long-term tax planning and the decumulation phase.
The moat is the rulebase as much as the solver: thirty years of encoded, continuously maintained federal and state tax, Social Security and benefit rules, carried under a regression suite re-run against every law change, plus patent-winning optimization algorithms built from economic theory rather than scraped text. The maintained surface is concrete: federal, Social Security, Medicare Part B and 42 state income tax codes, updated by the engineering team as provisions are released, on an annual law-update cycle. Stated plainly, because it will be checked: the solver is the replicable half — the mathematics is published, much of it by Kotlikoff himself. The rulebase is not, because encoding thirty years of law correctly is the decade.
Planning tools die on data entry. Inside SoFi the inputs problem disappears: Coach already aggregates across 12,000 institutions, and the member relationship spans banking, lending, investing and protection — which is the whole household balance sheet a lifetime optimization needs. Add the in-house financial planning team, delivering the same auditable plan the member sees in chat.
The instinct that AI is commoditizing software is correct, and it is the argument for this asset rather than against it.
What generative AI is rapidly commoditizing is interface, workflow, reporting and integration glue. None of that is what is on offer here. SoFi builds interface better than almost anyone; that is not the missing piece.
What AI does not produce is a validated rulebase or the evidentiary history that makes an output defensible. A model asked whether to pay down debt or fund a Roth will generate a fluent, confident, unverifiable answer. It has no correct reference point, so no error in it is decidable. MaxiFi’s is: rerun the engine and check.
Consider Intuit. Its enduring competitive advantage is not TurboTax’s interface or its AI features. Its moat is the tax-calculation engine. Large language models can generate plausible explanations, but they cannot reliably compute taxes, optimize outcomes, or produce audit-ready answers. Intuit can confidently deploy AI because every conversational interaction ultimately resolves against a deterministic rules engine designed to produce correct and defensible results.
The same principle applies to retirement and financial planning. Advisors and consumers will interact through increasingly sophisticated AI interfaces, but the value will reside in the analytical infrastructure beneath them. The AI asks the questions. The rules engine produces the correctly computed answer.
Every planning incumbent has now attached generative AI to a goals-based engine — a language model in front of arithmetic that was never deterministic. As models improve they converge on one another, and the industry mistakes that agreement for accuracy. A perfect mimic of an approximation is still an approximation.
The part you would be buying is the part AI has made scarcer. MaxiFi does not approximate. It computes — iteratively, multivariately and simultaneously across taxes, benefits, longevity and cash flow, year by year for a whole life. It is provable, not merely confident: the answer that holds up when someone with an adverse interest checks the math.
And the clock is real. A build arrives in years; the member base, the agents and the category window run in quarters. Composer took weeks.
The report identifies, as explicit risks of agentic AI: auditability and transparency — complicated, multi-step agent reasoning can make outcomes difficult to trace or explain; domain knowledge — general-purpose agents may lack what complex, industry-specific tasks require; and autonomy — agents acting without human validation. FINRA and the U.S. Treasury have since published an AI Lexicon and a Financial Services AI Risk Management Framework.
The substance of a financial recommendation is governed regardless of the interface delivering it, and being “AI-generated” is not a liability shield. The exposure scales with the size of the advised population — and 14.7 million members, growing 35% a year, is a large denominator.
A correct-by-construction engine addresses the exposure directly: if the math is right, reproducible and auditable, the answer holds up on its own terms. And because the engine is deterministic, the assurance can be underwritten — a bounded accuracy guarantee no probabilistic rival can offer, because their output has no correct reference point to warrant.
It also starts from the defensible number: the most a household can safely spend with what it has, sustainable by construction — not an aspirational target that manufactures the wrong, litigable figure.
CBS MoneyWatch (May 7, 2026) ran an identical retirement question — a 50-year-old single woman retiring at 65 — through two leading AI models. The verdicts diverged. MIT’s Andrew Lo was quoted on the underlying structural point: today’s consumer AI carries no best-interest duty. Kotlikoff was quoted describing the risk that AI “may do more harm than good” when it mishandles claims like Social Security timing or substitutes an average for a maximum life expectancy.
A concrete, checkable example: AI engines trained before the One Big Beautiful Bill Act (enacted July 2025) told users the federal estate-tax exemption would “sunset” on January 1, 2026 — reverting to roughly half its level. In fact, the Act permanently raised the exemption to $15 million per person starting in 2026.
A model repeating pre-2025 training data would confidently tell a household to rush an irrevocable estate move it no longer needs — a costly, hard-to-reverse error delivered with total confidence. A computed engine, fed current law, does not carry stale assumptions forward as fact.
Neither example is about any single company’s brand. It is the same structural point twice: confidence is not correctness, and an answer’s value depends on the currency and correctness of the computation behind it — not the fluency of the sentence delivering it.
Larry’s Economics Matters Substack — 137,000+ subscribers — has run a six-post sequence testing named frontier engines against MaxiFi on dollar-specific household problems — the same questions Coach is being asked today. The variance across engines on identical, checkable prompts is the proof: the correctness cannot come from the model layer.
Acquiring MaxiFi acquires the megaphone these pieces ship from — pointed, with credibility no one in the category can match, at exactly the surface Coach now owns. The CBS finding is the named, neutral proof; the Substack series is the dated, dollar-specific record behind it.
Durable value accrues to whoever owns the deterministic engine under the trusted interface — not to the interface, and not to the model. SoFi has proven it understands this: Galileo, Technisys, Composer. The planning engine is the one layer left in consumer finance that nobody owns.
The brand promise is a correctness claim. Today it is delivered by a model that must disclaim itself. With a computed engine underneath, the promise becomes literal across 14.7 million members — and it is a claim Chime, Robinhood and the frontier assistants cannot truthfully make.
The claim persuades; the guarantee closes. MaxiFi’s determinism makes a planning-side accuracy guarantee offerable for the first time: a computational error is objectively decidable, so the warranty prices at a rounding error and is insurable. A Monte Carlo or rule-of-thumb competitor cannot offer it at any price.
A correct-by-construction engine retires the largest overhang on giving money advice to millions of members through a chat interface. We are not selling an insurance policy; the insurance is included. And there is exactly one MaxiFi — it will sit somewhere, and in the model layer it reaches every competitor through the same API you would rent.
A computed lifetime plan names the exact protection gap, the exact refinancing decision, the exact investment contribution — with a date and a dollar figure, derived rather than pitched. For an everything-app whose economics run on members taking a second and third product, that is the highest-intent surface in the business.
Acquire the computation, put the SoFi interface on it, ship it in weeks. The difference is that a lifetime plan is the highest-stakes question a member will ever ask — and the only one whose answer can be guaranteed.
MaxiFi is being offered through a focused strategic process — the engine, its IP, and thirty years of R&D. The preference is an acquisition; that is where the strategic value sits. Continuity de-risks it: Larry Kotlikoff intends to stay on with the acquirer in whatever capacity best serves the product — architect, spokesperson, advisor. The next step is a 30-minute live demonstration: MaxiFi solves a real household’s plan while the leading models are asked to match it. The gap is the thesis. Evidence deepens with commitment — nothing is deployed, nothing left behind, and the full case is provable in an acquirer’s first quarter of ownership.