The questions people actually ask about the NextGen UBE.

The exam itself

1.1What is the NextGen UBE?

It is the bar examination that began replacing the MBE-based Uniform Bar Exam in July 2026. It runs nine hours over a day and a half, is delivered on the examinee's own laptop through NCBE's secure platform at jurisdiction-managed sites, and reports on a 500 to 750 scale with each jurisdiction setting its own passing line. The deeper change is one of design: rather than testing memorized doctrine and inferring lawyering ability from it, the NextGen names the skills it means to measure and builds question types to reach them directly.

1.2What actually changed from the old MBE-based bar exam?

Three structural things, and they compound. Skills are scored directly, so the exam now reaches research judgment, counseling, negotiation, and client management. Provided law beats memorized law: the exam regularly hands you a statute, a rule, or an edited opinion and expects you to apply that text, including where it departs from the general rule you memorized, and it can test areas outside the memorized syllabus entirely by supplying all the law required. And the item families are new: a six-option select-two, an integrated question set built on a client file whose facts unfold as you work, and a legal research performance task with no ancestor on the old exam.

The subject scope narrowed at the same time. Taken together, that is why an outline stack assembled for the MBE prepares you for an exam you will never take, and why the published content scope for your administration, rather than anything you inherited, is the operative document.

1.3How long is the NextGen UBE, and how is the exam structured?

Nine hours over a day and a half, in three three-hour sessions, delivered on your own laptop through NCBE's secure platform at testing sites the jurisdictions manage. The first administration, in July 2026, ran across sixteen testing locations in ten jurisdictions. Three item families fill those nine hours: standalone multiple choice, integrated question sets, and performance tasks.

1.4What kinds of questions are on the NextGen UBE?

Standalone multiple choice takes roughly forty percent of exam time and arrives in two formats: the familiar one-of-four, and a six-option field from which you select exactly two. Integrated question sets take just over a quarter, and are the genuinely new family. Performance tasks account for roughly a third of the score and come in two species, the sixty-minute standard task and the legal research task. Each is treated at length in the formats section below.

1.5What subjects are tested on the NextGen UBE?

Eight areas carry the exam: Contracts, Evidence, Civil Procedure, Business Associations (which absorbs agency and partnership), Real Property, Torts, Constitutional Law, and Criminal Law and Procedure, which NCBE treats as a single area. Family law and trusts and estates enter the exam beginning July 2028, largely through resource-provided formats.

The more useful point is one most candidates miss: the published content scope is a closed list, and in places it is narrower than the outline you may have inherited. As published for the July 2026 through February 2027 administrations, the real property scope does not carry the rule against perpetuities, zoning and land use, or equitable conversion. The scope is published per administration and deserves to be read like the syllabus it is. Every doctrine keyed in the But For bank maps to an explicit line of it, and doctrines the outline does not list are rejected and logged, however tempting.

1.6Do I still have to memorize the law for the bar exam?

For most of the exam, yes: the eight scope areas are still tested from knowledge. What changed is that memory is no longer the whole game. The NextGen regularly supplies a statute, a rule, or an edited opinion and expects you to apply that provided text objectively, even when it cuts against the rule you learned, and it can reach subjects outside the memorized syllabus by providing every authority needed. Applying provided law against your own instincts is a distinct skill from recall, and it is where this exam separates scorers.

1.7What skills does the NextGen UBE say it is testing?

NCBE names them openly: issue spotting and analysis, legal research, legal writing, investigation and evaluation, counseling, negotiation and dispute resolution, and client relationship management. That list is not decoration, and most prep treats it as a table of contents rather than a specification. Weighing a settlement against a client's actual stated objectives, or judging whether a provided library even answers the client's question, has never been directly scored on a bar exam before.

1.8How is the NextGen UBE scored?

It reports on a 500 to 750 scale, with each jurisdiction setting its own passing line, and results have historically arrived months after the exam. Because July 2026 was the first administration in history, there is no published NextGen curve to reason from yet, and every claim about how hard the exam runs is currently a feeling calibrated against prep materials rather than against performance. That is exactly why But For's own difficulty estimates and score bands stay labeled provisional until real calibration data exists.

1.9When does the NextGen UBE reach my jurisdiction?

July 2026, the first administration, in ten jurisdictions: Connecticut, Guam, Idaho, Maryland, Missouri, the Northern Mariana Islands, Oregon, Palau, the U.S. Virgin Islands, and Washington.

July 2027 adds thirteen more: Arizona, Iowa, Kentucky, Minnesota, Nebraska, New Mexico, North Dakota, Oklahoma, South Dakota, Tennessee, Vermont, West Virginia, and Wyoming.

February 2028 brings Delaware, the District of Columbia, Illinois, and Montana, and is also the legacy UBE's final administration.

July 2028 is the largest wave: Alabama, Alaska, California, Colorado, Florida, Georgia, Hawaii, Indiana, Kansas, Maine, Massachusetts, Michigan, New Hampshire, New Jersey, New York, North Carolina, Ohio, Pennsylvania, Rhode Island, South Carolina, Texas, Utah, Virginia, and Wisconsin. That single date puts the majority of American bar candidates on the new exam. California's added state-specific component arrives no earlier than July 2029.

As of August 2026, Arkansas, Louisiana, Nevada, and Puerto Rico had not announced, and Mississippi had signaled adoption without a date. The dates above trace to NCBE's published adoption list; the picker on the home page covers all fifty-six jurisdictions. Confirm against your own admissions authority before planning around any date, including this one.

1.10What happened at the first NextGen bar exam administration?

Per NCBE's official statement, approximately 2,624 applicants were scheduled across ten jurisdictions and sixteen testing locations in late July 2026. Seven jurisdictions appear to have run without reported incident. Maryland had one site affected for under an hour and completed Day 1 on schedule. Missouri had site-specific network problems that delayed the start for many examinees and resumed the same day. Washington could not begin at all: 645 standard-time examinees at the Yakima venue were dismissed after hours of attempted fixes, and the state canceled the administration that evening. NCBE describes the failures as site-local and says they did not involve the NextGen platform itself.

The complete record, with every claim labeled by how well it is corroborated, is published in full here. Affected applicants should rely on official communications from their jurisdiction rather than on any news account, ours included.

The new formats

2.1What is a select-two question?

A multiple-choice item with six options, lettered A through F, from which you select exactly two. It suits questions with more than one right move: which two legal topics most need researching, which two facts carry the argument, which two authorities control. Its distractors are typically near-misses rather than obvious throwaways, which is the whole difficulty. It is also the format no legacy archive can supply, because it did not exist on the old exam. But For's bank holds 216 of them.

2.2How does partial credit work on select-two questions?

Partial credit is real. Selecting both correct options earns full credit; selecting one correct option earns partial credit; a blank earns nothing. The practical consequence is a rule worth internalizing long before exam day: never leave a select-two unanswered. A single confident selection is worth strictly more than silence, and no amount of uncertainty about the second option changes that arithmetic. Every select-two on the bench is scored on the same basis, so the arithmetic you practise against is the arithmetic the exam uses.

2.3What is an integrated question set?

One client matter, one role (a solo practitioner, a public defender, an associate), built on realistic documents: interview notes, a lease, a police report, a deposition excerpt, sometimes an excerpt of statute or an edited opinion. The set asks a mixed sequence of multiple choice, short answers, and medium-length written analysis, with facts revealed progressively as the matter develops. Some sets are doctrinal; others centre on drafting or editing a document, or on counseling a client whose stated objectives and constraints you are expected to actually use. This is the family with no MBE-era analogue at all, and the one candidates most often meet cold.

2.4Do earlier answers really lock inside an integrated set?

Yes. Once new information is revealed, the answers you have already given are fixed. That is the point of the format: it tests judgment on the facts then known, not the polished hindsight of a completed record, which is how the practice of law actually runs. But For's set runner enforces the same discipline, so "facts now known" is a mechanic on the bench rather than an honor system.

2.5What are the performance tasks on the NextGen UBE?

Two species, together roughly a third of the score. The standard task descends from the MPT: a file, a library of provided authorities, and one extended writing assignment, on a sixty-minute clock, with narrowing instructions and formatting rules that are themselves graded. The legal research task is new and gets its own answer below. But For ships six tasks across both formats, including a persuasive standard task, each on the real clock.

2.6What is the legal research performance task?

A task that tests research judgment itself rather than doctrine: four multiple-choice questions, a short answer, and a medium-length writing assignment. What it examines is authority discipline. Which source is primary and which secondary; which binds and which merely persuades; what a case actually held as against what it said in passing; and whether the library you were handed even answers the client's question. It is the skill the MBE never touched, which makes it the one most often underprepared, and the one where a few hours of deliberate practice moves the most.

2.7How much writing is on the NextGen UBE?

More than the format summary suggests. Performance tasks alone are roughly a third of the score, and the integrated sets add short and medium-length written answers on top of that. Written work is also the part of preparation most easily flattered by a study product that never scores it: reading a model answer and nodding is not the same act as producing one against a clock and being told, with evidence, which points you actually earned.

2.8Are the NextGen multiple-choice questions basically MBE questions?

Not for long. Early administrations deliberately resemble MBE-style questions, and the format variety grows from there. Even among the one-of-four items, many now pose a lawyering question rather than a pure doctrinal call: what to research first, what to advise, which fact changes the analysis. Treating the multiple-choice portion as a familiar quantity because the option count looks familiar is one of the more common ways to underprepare for this exam.

Preparing for it

3.1When should I start preparing for the NextGen UBE?

There is no sourced answer, and the tidy figures that circulate (a fixed number of weeks, a fixed number of hours) come from marketing rather than from any study of this exam, which has a single administration behind it. Publishing one would violate our own rules.

What can be said without guessing: the formats, not the doctrine, are the part that takes time to internalize, because select-twos and integrated sets punish habits the MBE rewarded, and any record of your own performance needs volume behind it before it means anything. Access runs through the February 2028 administration, so the bench fits whatever calendar you and your jurisdiction are actually on.

3.2Can I use But For alongside a full-service bar course?

Yes, and that is the design. But For is the drilling layer and nothing else: no videos, no live classes, no tutors. It is built to sit beside whatever outlines, lectures, or commercial course you already use, and to supply the one thing an MBE-era archive cannot, which is practice in the formats the current exam actually delivers.

3.3Is old MBE practice still worth doing?

Partly, and it is worth being precise about which part. On the doctrinal subjects the two exams share, a well-written legacy question still exercises real rule knowledge, and rule knowledge has not stopped mattering. What a legacy question cannot do is touch the select-two, the integrated set, or the research task, and it will spend some of your hours on doctrines the current content scope no longer lists. So legacy multiple choice is a reasonable doctrine exercise and a poor exam rehearsal. Treat it as the former and you will not be misled. That framing is our judgment, offered as such.

3.4What is the right way to use practice questions?

Four principles, each chosen because it survives scrutiny rather than because it sells.

Retrieve rather than reread. Answering under exam conditions is the act that builds durable memory; the testing effect is among the most replicated findings in cognitive psychology. Rereading an outline feels productive and mostly is not.

Practise in the exam's own grammar. Skills transfer best to the conditions they were rehearsed in, so exam day should run on patterns already laid down rather than on translation from a format that no longer exists.

Interleave. Mixed-subject lanes force the harder discrimination ("which doctrine is this?") that blocked, one-subject drilling never asks. It feels worse and works better; that trade is deliberate.

Read the feedback, not the score. Every option should explain why it seduces or fails, and every item should leave you one line you will actually retrieve in the room. And work against a clock: nine hours punishes perfectionism, so knowing when a point is earned and it is time to move is itself a scored skill.

3.5What should I do about a subject I keep missing?

Rebuild it from your own record rather than from the outline. On the bench, the ledger tracks accuracy by subject and separates the misses you were confident about, and one tap assembles a gap-training session: your confident misses retried first, then unseen items from your two weakest subjects. The reason to start with the confident misses is that they are precisely the errors an outline reread will never find, because they do not feel like gaps.

3.6Why record confidence on every answer?

Because the most dangerous error class is the confident miss. A question you knew you were guessing at is already on your list. A question you were sure about and got wrong is not, and nothing in a raw score surfaces it. Capturing Sure or Unsure on every answer makes miscalibration visible, then measurable, then trainable. The aim is an examinee whose confidence is information.

3.7How do I read my own practice results?

Look at four things rather than one. Confidence-split accuracy separates what you know from what you merely recognize. Pace against each question's time target matters because being right slowly is a scoring problem. Accuracy across difficulty tiers (the bench runs three: Foundation, Administration, Stretch) tells you whether your ceiling or your floor needs the work. And the calibration gap, the distance between how sure you were and how right you were, is the number most likely to change how you study.

What none of it supplies is a prediction of your bar result. A practice record describes practice.

3.8What is the Readiness Benchmark, and what does a band mean?

A timed cross-tier assembly with one AI-graded written component, ending in a provisional band and four dials, re-runnable weekly with the trend kept on your dashboard. It is built to measure movement over time, which is the one thing a single sitting cannot show. The band is labeled provisional and stays labeled provisional until real calibration data earns the label's removal. It is not a pass prediction, and it will never be presented as one.

But For, specifically

4.1What is But For?

An independent study product and publication for the NextGen UBE. The study product is a drilling bench built format for format to the real exam: original multiple choice in both NextGen formats including six-option select-twos scored with real partial credit, integrated question sets where the facts unfold and earlier answers lock, both performance-task formats on a real clock, AI grading of written work that quotes its evidence, and a ledger that tracks confidence on every answer and turns your confident-wrong record into gap training.

The publication is the sourced record of how this exam behaves in the world, starting with the first administration, every claim labeled by how well it is corroborated. The editorial standard is published in full so you can hold us to it.

4.2What exactly is included for $199?

The complete bank as it stands and as it grows: 534 original questions in both formats, comprising 216 select-twos with real partial credit and 318 select-one items, plus all twelve integrated question sets and all six performance tasks. Also the free 90-minute practice session and every session assembled through v1; rubric-transparent AI grading on written answers; the ledger with gap training; the Readiness Benchmark; the first-administration Calibration Memo; a permanent Founding Member mark on the account; and every content update through the February 2028 administration. One payment. Nothing recurs.

4.3What is But For not?

No videos, no live classes, no tutors, no lecture series. It is the doing layer and nothing else, built to sit beside whatever outlines you already use. It is also not legal advice, and it does not promise a score. If what you need is someone to teach you contracts from zero, this is not that product, and we would rather say so here than after you have paid.

4.4How long does But For access last, and is it a subscription?

Access runs through the February 2028 UBE administration, the one at which the MBE era ends entirely, unless a different period is stated at checkout. It is not a subscription: one payment, nothing recurring, no renewal to forget to cancel. Every content update shipped during your access period is included, which for a product built to an exam still finding its shape is most of the value.

4.5What happens after I pay?

Checkout runs through Stripe, so But For never receives or stores your full card details. You then sign in with the purchase email using a one-tap link, with no password to set. The bench opens immediately: the bank, the sets, the tasks, grading, and the ledger. It is not a preorder, and nothing is staged behind a release schedule.

4.6How does the AI grading work, and can I overrule it?

Every written answer is graded against the task's rubric or representative answers, point by point, and the grader quotes your own words as the evidence for each point it awards or refuses. You can watch it work on a real answer before buying anything.

It is deliberately assistive rather than authoritative. If you disagree, you overrule it, and your call stands on the record. That design is not modesty about the technology; it is the only arrangement that makes the grade useful. A grader that cannot be questioned is a grader you cannot learn from, and a grade that will not show its evidence is a number pretending to be feedback.

4.7Is the content original, and do you sell predicted questions?

Every scenario, statute, and option is written in-house in our fictional State of Meridian and similarity-gated against NCBE's official samples before it ships. Nothing licensed, nothing farmed, nothing echoed.

There are no predicted questions, and there never will be. Examinees are bound to confidentiality about what the exam asked, and we hold ourselves to the same line voluntarily and absolutely: we do not solicit, record, or repeat exam content. What we publish is examinee experience, meaning logistics, pacing, and format mechanics. Anyone offering you the questions is offering you something they should not have.

4.8How are the questions written and checked before they ship?

Every batch passes six scripted gates, each one a hard stop. Structure: exact format integrity, with a teaching explanation for every option and a one-line discriminator worth remembering. Doctrine audit: the batch is diffed against a regenerated map of everything already in the bank, so coverage deepens instead of repeating. Naming: every coined name is checked against the entire corpus and collisions are renamed. Key balance: answer letters are distributed and audited, so patterns cannot be gamed. Import acceptance: the full corpus re-imports with zero errors, or the batch is not done. Scope citation: every keyed doctrine maps to an explicit line of NCBE's published content scope, verified against the source.

Where jurisdictions genuinely divide, the materials flag the split rather than pretending the majority rule is the only rule. Authorship is in-house to a law-faculty standard under one editorial direction, and every batch lands as a draft that must survive founder review before members see it.

4.9What happens if I think a question is wrong?

Write to support@butfor.ai. Every report is reviewed and verified errors are corrected. The law changes and reasonable minds differ; we work to a high accuracy standard, flag genuine jurisdictional splits, and label corrections to published material as corrections rather than editing them out of history.

4.10Who is behind But For, and is it affiliated with the NCBE?

But For is an independent publication and study product of Ink Forms Rivers, doing business as But For. It is not affiliated with, sponsored by, or endorsed by the National Conference of Bar Examiners or any bar examining authority; "NextGen UBE" is referenced solely to identify the examination, and those marks are NCBE's property.

Items are authored and edited in-house to a law-faculty standard under one editorial direction. Independence is the point rather than a disclaimer: every structural claim we make about the exam traces to NCBE's published materials, and where we are inferring, we say so.

4.11Can I try But For before buying?

Yes, and without an account. A three-question calibration runs live on the home page in the exam's real formats, with per-option explanations and the confidence capture intact. The live grader will grade one written answer per email address, holding the same rubric line members see. And the first-administration report is free. None of it is a demo build; it is the product, with the paywall moved.

4.12Does But For guarantee a passing score?

No, and no honest study product can. Bar results depend on many factors outside any preparation tool. So there are no pass-rate claims here, no outcome guarantees, no countdown timers or fake seat counters, and no predicted questions; score bands and difficulty estimates stay marked provisional until data earns the label's removal. The claim we do make is narrower and stronger: the mechanisms are the ones the evidence supports, aimed at the skills the examiners publish, drilled in the format they deliver.

Practical and administrative

5.1What do I need to use But For?

A browser, on any device. There is nothing to install and no separate app. One practical note: because the exam itself is delivered on your own laptop, practising written work at a keyboard is closer to exam conditions than practising it on paper.

5.2Can I share an account with my study group?

No. Accounts are personal and may not be shared, pooled, or resold, and the content license is limited, personal, and non-transferable. Study groups need individual licenses. The content may also not be scraped, bulk-downloaded, or used to train, fine-tune, or evaluate any machine-learning system. A single item may be screenshotted for public commentary if it is attributed.

5.3What data does But For keep about me, and what is done with it?

The minimum needed to run a study platform: your email address, your purchase record via Stripe (never your full card number), and your practice data, meaning answers, confidence tags, error tags, timing, and scores. That data runs the product, powers transactional email, and improves item calibration in aggregated, de-identified form. Personal data is not sold, no advertising trackers run on the site, and measurement uses cookieless analytics. Account and practice data are kept while the account is active and for 24 months after, then deleted or de-identified. The full privacy policy is short and worth reading.

5.4Can I get a copy of my record, or have it deleted?

Yes. Email support@butfor.ai to access, correct, export, or delete your personal data. Requests are honored within 30 days, verified against your account email. The substance of access, deletion, and portability requests is honored for everyone regardless of where you live, whether or not a statute in your jurisdiction happens to require it.

5.5Will my performance data ever be shared with a bar examiner, law school, or employer?

No. Individual performance data is never shared with schools, employers, or examiners absent your explicit request or genuine legal compulsion. Your practice record exists to train you, and nobody is grading you on it but you. That matters more than it sounds: a record you are afraid of is a record you will not answer honestly, and a dishonest confidence tag is worse than none.

5.6Who is But For for?

Law graduates and law students preparing for the bar examination, who must be at least 18 and able to form a binding contract. Practically, it is built for anyone taking a NextGen administration who wants practice in the exam's real formats rather than a retrofitted archive, whether that administration is the next one or one in 2028.

5.7How do I reach a human at But For?

support@butfor.ai reaches a human who reads every message. That address handles questions about the product, error reports on individual items, data requests, and account access. If you have taken a NextGen administration, it is also where firsthand experience accounts are collected: logistics, pacing, format mechanics, never exam content.

Question not here? support@butfor.ai · we add the ones people actually ask.